Machine learning constructs a diagnostic prediction model for gangrenous perforation of acute appendicitis in elderly patients

Machine learning constructs a diagnostic prediction model for gangrenous perforation of acute appendicitis in elderly patients | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Machine learning constructs a diagnostic prediction model for gangrenous perforation of acute appendicitis in elderly patients Jie Cheng, Yanbo Sun, Bin Yang, Yun Gong, Die Fan, Yongqilin Shuang, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7656285/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 24 Apr, 2026 Read the published version in BMC Surgery → Version 1 posted 13 You are reading this latest preprint version Abstract Background As life expectancy rises and elderly populations grow, acute appendicitis incidence increases, often manifesting with nonspecific symptoms that challenge diagnosis. This study applied machine learning techniques to build a predictive model for gangrenous perforation, examining clinical features and risk factors in elderly patients with acute appendicitis. Methods We conducted a retrospective analysis of elderly patients undergoing laparoscopic appendectomy for acute appendicitis at The Second Affiliated Hospital of Kunming Medical University, China, from June 2021 to January 2024 (n = 251). Patients were classified into gangrenous perforation (n = 69) and non-gangrenous (n = 182) groups, then randomly split into training (70%) and test (30%) sets. Univariate analyses, including t-tests, Spearman correlations, and chi-square tests, assessed differences across 38 variables in both sets. The least absolute shrinkage and selection operator (LASSO) screened features from the training set, informing models via logistic regression (LR), extreme gradient boosting (XGBoost), support vector machines (SVM), and random forest (RF). Model performance was evaluated using area under the receiver operating characteristic curve (AUC), with decision curve analysis assessing clinical applicability. Results In the training set, RF yielded the highest AUC (0.999, 95% CI: 0.998–1.000), followed by XGBoost (0.975, 95% CI: 0.953–0.998), LR (0.774, 95% CI: 0.692–0.856), and SVM (0.768, 95% CI: 0.684–0.852). In the test set, LR performed best (AUC 0.768, 95% CI: 0.642–0.893), surpassing SVM (0.751, 95% CI: 0.620–0.882), XGBoost (0.725, 95% CI: 0.584–0.867), and RF (0.686, 95% CI: 0.533–0.839). LR also showed the highest accuracy (0.784) and specificity (0.741), while XGBoost had the greatest sensitivity (0.836). Conclusions Among the models, LR emerged as the most effective for predicting gangrenous perforation in elderly acute appendicitis patients, offering robust accuracy and reliability. Its nomogram provides a noninvasive aid for clinical diagnosis. Machine learning gangrenous perforation acute appendicitis elderly patients diagnosis Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 1. Introduction Aging is advancing at an unprecedented rate all over the world as evidenced by World Health Organization that the global population of people aged 60 and above is expected to double, reaching approximately 2.1 billion by 2050, and the proportion of the population aged 60 and above will increase from 13.5% in 2020 to 22% in 2050, and this rising trend has a profound impact on the different aspects of society particularly the healthcare sector [ 1 , 2 ]. The incidence rates of various geriatric diseases like acute appendicitis in the elderly patients is particularly fluctuating with a significant upward trend of 3.2% has been reported annually [ 3 ]. These vigorous rising facts pretend the higher risk of acute abdominal diseases in the elderly population. With the rise in acute appendicitis in the elderly the condition also differs from young patient due to unique pathophysiological characteristics. A major concern in clinical treatment is high incidence of gangrene and perforation (30%-40%) which is 2–3 times higher than young patients and poses significant challenges like as high risk of postoperative complications, prolonged hospital stay and hereby increased medical costs [ 4 – 6 ]. The poor prognosis in elderly patients with acute appendicitis is due to multiple factors like weak immune system, degenerative changes in blood vessels, and alteration in pain sensitivity. As immune response declines, the body's defense mechanisms become less responsive to inflammation, allowing appendiceal inflammation to progress quickly and making it difficult to control effectively. [ 7 ]. The degenerative changes in blood vessels affect the blood supply of the appendix resulting in local tissue ischemia and hypoxia, thereby increasing the risk of gangrene and perforation [ 8 ]. The decrease in pain sensitivity is also a factor that cannot be ignored. Similarly, the reduction in pain perception ability of elderly patients often worsen the condition due to delay in detection of pain symptoms in a timely manner at the initial stage of appendicitis [ 8 ]. Thus, the clinical diagnosis and treatment become challenging in elderly population making them a high-risk group in the diagnosis and treatment of acute abdominal diseases. Another challenge is early and accurate diagnosis however, complexity of underlying diseases, decline in body functions with age, hidden or atypical clinical symptoms limits the use of traditional diagnostic methods [ 9 ]. The typical migratory abdominal pain, which is a hallmark symptom of acute appendicitis, however, decrease in pain sensitivity and changes in nerve conduction often leads to absence of such typical symptoms in most elderly patients [ 8 , 10 ]. Similarly, peritoneal irritation being a significant base for intra-abdominal inflammation is often not experienced by elderly patients (~ 28% patients) [ 11 ]. Even loose abdominal wall muscles and slow response also results in poor diagnosis or misdiagnosis [ 12 ]. Irrespective of this, use to commonlaboratory tests, like as C-reactive protein (CRP) and white blood cell count (WBC) are offers a for diagnosis is limited due to inconsistent sensitivity and specificity of CRP and wide range of WBC, and also interfered with by other factors in elderly patients, resulting in the non-specific alterations in these indicators, hereby making it difficult to accurately to independently diagnose the gangrene and perforation [ 13 , 14 ]. Nevertheless, the imaging techniques like ultrasonography, computed tomography (CT), and magnetic resonance imaging (MRI) offers a best aid in diagnosis of acute appendicitis in the elderly patients. But intestinal gas, position of appendix, and increased adipose tissue compromise the sensitivity of non-invasive ultrasonogphy in elderly patients as compared to young patients (68% vs., 92%) [ 15 – 17 ]. While, CT offers best alternate for ambiguous diagnosis and increased the sensitivity to 89%, for elderly patients with poor renal function but the use of contrast agents has been reported to even cause the renal failure in severe cases [ 18 ]. Nonetheless, MRI being a radiation-free imaging method has shown a high accuracy in diagnosing the complex appendicitis but the cost and time and limited access hinder wide spread use in the diagnosis of appendicitis hereby necessitating the use of alternatives [ 19 ]. The rapid development of medical technology has enabled the machine learning (ML) to play an increasingly important role in the clinics, and is helpful in accurate diagnosis, and formulation and optimization of patient-based treatment plans for different diseases. Evidences from breast cancer diagnosis and classification reported that ML has achieved remarkable achievements in various medical imaging modalities such as mammography, ultrasound, MRI, histology, and thermography with high diagnostic accuracy [ 20 ]. It has improved the precision of decision-making and patients’ treatment outcomes in clinics, thus offering a great potential in clinical diagnosis, personalized treatment, and health monitoring [ 21 ]. Nevertheless, the application of ML technology in diagnosis of gangrene and perforation of acute appendicitis in the elderly is still in the initial exploration stage with limited reports. Thus, in this study, we utilized the ML technology to construct a diagnostic prediction model of gangrene and perforation of acute appendicitis in the elderly patients, hereby focusing on in-depth analysis of the clinical characteristics of gangrene and perforation of acute appendicitis to provide efficient and reliable auxiliary diagnostic tools for gastrointestinal surgeons. 2 Material and Methods A retrospective study from June 2021 to January 2024 was carried out at the Second Affiliated Hospital of Kunming Medical University, China. The research protocol and consent to participate was approved by the ethical Review Committee of The Second Affiliated Hospital of Kunming Medical University, China. 2.1 Patients’ selection The inclusion criteria for patients was as: the patients were diagnosed with acute appendicitis through a comprehensive diagnosis based on pre-operative classic clinical manifestations (migratory right lower quadrant pain, tenderness and rebound tenderness at McBurney's point), imaging features (abdominal CT showing thickening of the appendix and exudation of surrounding fat streaks), and laboratory indicators (elevated white blood cell count), and were confirmed by laparoscopic surgical resection combined with intraoperative exploration and postoperative pathological examination; having age ≥ 60 years old with complete perioperative data, including pre-operative CT imaging data, laboratory tests (complete blood count, comprehensive biochemistry, coagulation function, electrolytes), and postoperative pathological reports. The patients with following complications like active infection in other parts, history of malignant tumor or postoperative pathological examination suggesting a neoplastic lesion, abnormal coagulation function or use of anticoagulant/antiplatelet drugs within 30 days before surgery, hematological diseases, and severe hepatic and renal insufficiency were not included in the study. 2.2 Parameters of study The data collected was categorized into baseline data, per-operative laboratory indicators, and imaging features and pathological and intraoperative records. Briefly the baseline data comprised of demographic characteristics (age, gender, weight, height, BMI), underlying diseases (diabetes, hypertension), lifestyle habits (smoking, drinking), symptomatic characteristics (migratory right lower quadrant pain, nausea and vomiting, anorexia, fever, tenderness and rebound tenderness in the right lower quadrant), and disease course-related indicators (time from onset to surgery, use of preoperative antibacterial drugs). Thepre-operative laboratory indicators included inflammatory markers (white blood cell count WBC, C-reactive protein CRP, procalcitonin PCT), hematological parameters (platelet PLT, hemoglobin HGB, hematocrit HCT, red blood cell distribution width RDW), metabolic indicators (albumin Alb, serum Na+, direct/indirect bilirubin DBIL/IBIL), and coagulation function (D-D dimer), and the pre-operative imaging features were the minimum diameter of the appendix, thickening of the appendiceal wall (defined as thickening when > 3mm), the presence of fecaliths in the appendiceal lumen, and the presence of surrounding fluid. While, the pathological and intraoperative records withhold verifying the accuracy of the clinical diagnosis based on the postoperative pathological diagnosis and surgical exploration results. The data were extracted through a structured electronic medical record system to ensure the integrity and consistency of the information, providing comprehensive data support for subsequent multivariate analysis. 2.3 Data processing, and construction of ML prediction models After completing the clinical data collection in the preliminary stage of the study, the integrity and accuracy of the data were first checked. Records with obvious logical errors were removed, and the missing values were filled using K-nearest neighbors (KNN) algorithm [ 22 ] which screens the nearest neighbor samples by calculating the Euclidean distance between samples, and performs data imputation based on the similarity features. For dataset splitting, a fixed random seed strategy was adopted by dividing the total dataset into a training set (70%) and a test set (30%) at a ratio of 7:3. Lasso regression (Least Absolute Shrinkage and Selection Operator) was applied for high-dimensional feature selection on the data of the training set, which introduced the L1 regularization constraint condition to compress the coefficients of redundant features to zero, and finally selected a feature subset that was significantly correlated with the target variable. The ML prediction models comprised of logistic Regression (LR) with odd ration (OR), XGBoost, support vector machine (SVM) with selection of kernel functions (such as the Gaussian kernel and the polynomial kernel), and random forest (RF) models as reported previously [ 23 – 27 ]. A multi-dimensional verification strategy was adopted to systematically evaluate the generalization ability and clinical applicability of the prediction model. Firstly, we constructed the Receiver Operating Characteristic (ROC) curve to quantify the model's ability to distinguish gangrene and perforation events, and calculated the area under the curve (AUC) as the core evaluation index. In addition, we also supplemented it with classification performance indicators such as accuracy, sensitivity, and specificity. The DeLong non-parametric test method [ 28 ] was used for the difference test of the ROC curves among the models. Then, we visualized the consistency between the model's predicted probability and the actual observed probability through the calibration curve, and combined the Hosmer-Lem show test to quantify the goodness of calibration (P > 0.05 indicates that the model is well calibrated). This calibration curve shows the degree of coincidence between the predicted risk and the true risk distribution in the form of quantile grouping. Finally, in the evaluation of the degree of risk and benefit, we used Decision Curve Analysis (DCA) to quantify the clinical net benefit of the model under different risk thresholds, to evaluate the practical application value of the model. 2.4 Statistical analysis This study systematically analyzed 38 clinical characteristics of 251 patients based on the R statistical platform (version number 4.4.2; access website: https://www.R-project.org ). In the data preprocessing stage, firstly, the Kolmogorov-Smirnov test was used to evaluate the normal distribution characteristics of continuous variables: for the indicators that conform to the normal distribution, the independent samples T-test was adopted, and the results were presented as mean ± standard deviation [ ]. For the data with non-normal distribution, the Mann-Whitney rank sum test was used, and it was described by the median and interquartile range (first quartile, third quartile) [M(P25, P75)]. Categorical variables were analyzed by the chi-square test, and the frequency distribution was expressed as the number of cases and the proportion [N(%)]. All statistical analyses used a two-sided P value < 0.05 as the criterion for judging that the difference was statistically significant. The R packages used were: 'caret', 'tidyverse', 'autoReg', 'ggplot2', 'compareGroups', 'Table 1 ', 'plyr', 'corrplot', 'glmnet', 'rrtable', 'Hmisc','reportROC', 'rmda', 'randomForest', 'dplyr', 'rms', 'data.table', 'xgboost', 'ggpubr', 'Matrix', 'e1071', 'nortest', 'plotly', 'Ckmeans.1d.dp', 'ggprism', 'CBCgrps', 'DiagrammeR','shapviz', and 'pROC'. Table 1 Comparison results of general clinical characteristics on two groups Characteristics Training Set (n = 177) Test Set (n = 74) Non-gangrene and Perforation Gangrene and Perforation P-value Non-gangrene and Perforation Gangrene and Perforation P-value Patients (n) 128 49 54 20 Age (years) (IQR) 68 (64,73.25) 68 (64,75) 0.654 68 (64,71) 68.5 (64.75,74.25) 0.241 Weight (kg) 60.47 ± 8.97 57.55 ± 9.78 0.060 59.74 ± 10.23 61.25 ± 11.96 0.62 Height (cm) 161.09 ± 7.65 160.76 ± 7.01 0.788 160.81 ± 9.02 161.4 ± 5.29 0.731 BMI (kg/m 2 ) 23.32 ± 3.25 22.2 ± 3.03 0.034 23.09 ± 3.57 23.56 ± 4.70 0.645 Gender (n, %) Female 72(56) 22(45) 0.236 30(56) 9(45) 0.585 Male 56(44) 27(55) 24(44) 11(55) Diabetes (n, %) No 115(90) 45(92) 0.783 47(87) 18(90) 1 Yes 13(10) 4(8) 7(13) 2(10) Hypertension (n, %) No 80 (62) 35 (71) 0.348 37 (69) 9 (45) 0.113 Yes 48 (38) 14 (29) 17 (31) 11 (55) Smoking (n, %) No 117(91) 37(76) 0.01 50(93) 16(80) 0.2 Yes 11(9) 12(24) 4(7) 4(20) Drinking (n, %) No 120(94) 45(92) 0.74 53(98) 18(90) 0.176 Yes 8(6) 4(8) 1(2) 2(10) Time from Onset to Surgery (d) (IQR) 1(1,2) 1(1,3) 0.12 1(1,2) 1(1,2) 0.781 Use of Antibacterial Drugs (n, %) No 70(55) 23(47) 0.45 27(50) 10(50) 1 Yes 58(45) 26(53) 27(50) 10(50) Migratory Right Lower Quadrant Pain (n, %) Yes 128(100) 49(100) 1 54(100) 20(100) 1 Nausea and Vomiting (n, %) No 64(50) 28(57) 0.495 31(57) 9(45) 0.491 Yes 64(50) 21(43) 23(43) 11(55) Anorexia (n, %) No 28(22) 7(14) 0.356 12(22) 2(10) 0.326 Yes 100(78) 42(86) 42(78) 18(90) Fever (n, %) No 114(89) 32(65) < 0.001 49(91) 15(75) 0.122 Yes 14(11) 17(35) 5(9) 5(25) Tenderness in Right Lower Quadrant (n, %) No 7(5) 1(2) 0.447 4(7) 2(10) 0.659 Yes 121(95) 48(98) 50(93) 18(90) Rebound Tenderness in Right Lower Quadrant (n, %) No 33(26) 12(24) 1 18(33) 5(25) 0.685 Yes 95(74) 37(76) 36(67) 15(75) 3 Results 3.1 General clinical characteristics A total of n = 251 elderly patients who underwent laparoscopic appendectomy were included as per selection criteria and the general clinical characteristics of the research subjects are presented in Table 1 . The patients (n = 251) with acute appendicitis were divided into the gangrene and perforation group (n = 69) and the non-gangrene and perforation group (n = 182). Among them, the training set (n = 177) included n = 49 patients with gangrene and perforation and n = 128 patients without gangrene and perforation; the test set (n = 74) included n = 20 patients with gangrene and perforation and n = 54 patients without gangrene and perforation. The results showed that in the training sample set of this study, the gender distribution in the gangrene and perforation group for male and female was 55% (n = 27) and 45% (n = 22), respectively. The average age of the patients was 68 years old. In terms of comorbidities, history of diabetes, hypertension, smoking, and drinking was ,8% (n = 4), 29% (n = 14), 24% (n = 12), and 8% (n = 4), respectively. The average time from the onset of the disease to the operation was 1 day, while 53% (n = 26) patients had a history of taking antibacterial drugs before the onset of the disease. All patients showed migratory right lower quadrant pain. A total of 43% (n = 21) patients had symptoms of nausea and vomiting, 86% (n = 42) patients had anorexia, 35%(n = 17) patients had a history of fever, 98%(n = 48) patients had tenderness in the right lower quadrant, and 76% (n = 37) patients had rebound tenderness in the right lower quadrant (Table 1 )While, in the non-gangrene and perforation group, there were 44%(n = 56) male and 56% (n = 72)female patients. The average age of the patients was 68 years old. Patients with history of diabetes, hypertension, smoking, and drinking were 10% (n = 13), 38% (n = 48), 9% (n = 11) and 7% (12), respectively. The average time from the onset of the disease to the operation was 1 day; while 45% (n = 58) patients had a history of taking antibacterial drugs. All patients had migratory right lower quadrant pain. A total of 50% (n = 64) patients with nausea and vomiting78%(n = 100) patients with anorexia, 11% (n = 14) patients with a history of fever, 95% (n = 121) patients with tenderness in the right lower quadrant, and 74% (n = 95) patients with rebound tenderness in the right lower quadrant were identified (Table 1 ). In the training set, there were significant statistical differences in BMI (P = 0.034), smoking history (P = 0.01), and fever history (P < 0.001) between the gangrene and perforation group and the non-gangrene and perforation group; while in the test set, no statistically significant differences in characteristics were found between the gangrene and perforation group and the non-gangrene and perforation group. 3.2 Preoperative laboratory tests The results of preoperative laboratory tests (Table 2 ) showed that In training set there were significant differences between the gangrene and perforation group and non-gangrene and perforation group in terms of WBCs (P = 0.001), neutrophils (P = 0.004), lymphocytes (P < 0.001), CRP (P < 0.001), procalcitonin (P < 0.001), D-dimer (P < 0.001), indirect bilirubin (P = 0.03), Alb (P < 0.001), and serum Na⁺ level (P = 0.012). While, in the test set, the WBC (P = 0.016), neutrophils (P = 0.027), lymphocytes (P = 0.014), hemoglobin (P = 0.036), hematocrit (P = 0.02), CRP (P < 0.001), procalcitonin (P = 0.009), D - dimer (P = 0.032), and Alb (P = 0.037). Table 2 Comparison results of preoperative laboratory tests on two groups Characteristics Training Set (n = 177) Test Set (n = 74) Non-gangrene and Perforation Gangrene and Perforation P Non-gangrene and Perforation Gangrene and Perforation P Patients (n) 128 49 54 20 PLT (*10^9/L) 217.16 ± 65.51 203.33 ± 54.16 0.155 200.31 ± 54.02 201.65 ± 72.74 0.941 WBC (*10^9/L) 11.18 ± 3.85 13.64 ± 4.56 0.001 11.98 ± 3.18 14.36 ± 4.82 0.016 NEUT (%)(IQR) 83.65 (73.65,88.73) 87(82.6,91) 0.004 83.85(78.1,87.2) 86.55 (83.03,91.7) 0.027 LYM (%)(IQR) 11.5(7.25,19.7) 8.3(4.9,11.5) < 0.001 11.9(7.53,15.55) 7.9(4.47,11.03) 0.014 HGB (g/L) 144.62 ± 14.69 146.92 ± 19.68 0.46 146.72 ± 14.82 137.75 ± 18.95 0.036 HCT (%)(IQR) 0.43(0.41,0.46) 0.43(0.4,0.48) 0.51 0.44(0.42,0.46) 0.41(0.39,0.44) 0.02 MCV (fL) 92.08 ± 4.34 92.66 ± 4.59 0.448 92.19 ± 4.63 90.48 ± 6.33 0.205 MCHC (g/L)(IQR) 337 (331.75,342) 339(328,347) 0.585 334.5 (331.25,340) 334.5 (330.75,343) 0.63 RDW (fL) 44.03 ± 3.02 44.27 ± 3.42 0.648 43.62 ± 2.53 44.24 ± 3.04 0.417 MPV (fL)(IQR) 9.9(9.17,10.7) 10(9.4,10.4) 0.887 10.4(9.8,11.38) 10(9.35,11) 0.281 PCT (µg/L)(IQR) 0.11(0.06,0.32) 0.84(0.19,7.62) < 0.001 0.09(0.06,0.7) 1.41(0.08,40.73) 0.009 CRP (mg/L)(IQR) 42.34 (13.59,70.15) 91.39 (44.21,142.11) < 0.001 33.14 (10.8,63.51) 98.88 (36.4,136.29) < 0.001 D-Dimer (µg/L)(IQR) 0.51(0.34,0.99) 1.08(0.66,1.98) < 0.001 0.54(0.39,1.13) 1.61(0.55,5.11) 0.032 DBIL(µmol/L)(IQR) 4.9(3.48,7) 5.5(3.6,8.1) 0.231 4.3(2.92,6.05) 5.7(3.33,8.75) 0.154 IBIL(µmol/L)(IQR) 15.25 (11.1,22.45) 19.9(12.9,27.9) 0.03 16.95(10.38,23) 15.15 (11.78,19.62) 0.831 ALB(g/L)(IQR) 40.65 (38.48,42.73) 38.6(35.7,41.2) < 0.001 40.55 (38.6,42.42) 38.8 (35.77,40.62) 0.037 Na + (mmol/L) 136.09 ± 3.10 132.65 ± 14.53 0.012 136.35 ± 2.43 134.86 ± 3.65 0.104 Abbreviations: CRP = C-reactive proteins; WBCs = white blood cells 3.3 Preoperative imaging examinations In terms of imaging examinations (Table 3 ) there were statistically significant differences in all characteristics between the non-gangrene and perforation group and the gangrene and perforation group in the training set. While, only the minimum diameter of the appendix (P = 0.001) in the test set, showed significant difference between the non-gangrene and perforation group and the gangrene and perforation group. Table 3 Comparison results of preoperative imaging examinations on two groups Characteristics Training Set (n = 177) Test Set (n = 74) Non-gangrene and Perforation Gangrene and Perforation P-value Non-gangrene and Perforation Gangrene and Perforation P-value Patients (n) 128 49 54 20 Minimum diameter of the appendix (cm) (IQR) 0.9(0.7,1) 0.9(0.8,1.1) 0.04 0.8(0.76,1) 1.2(0.98,1.25) 0.001 Appendiceal wall thickening (n, %) 0.004 0.569 No 18(14) 0(0) 4(7) 0(0) Yes 110(86) 49(100) 50(93) 20(100) Fecalith in the appendiceal lumen (n, %) 0.049 1 No 66(52) 34(69) 31(57) 11(55) Yes 62(48) 15(31) 23(43) 9(45) Fluid around the appendix (n, %) 0.025 0.207 No 28(22) 3(6) 14(26) 2(10) Yes 100(78) 46(94) 40(74) 18(90) 3.4 Prediction potential of four machine learning models We next evaluated the prediction potential of four ML models and results indicated that during the training stage the AUC value of RF reached 0.999 (95% CI, 0.998–1.000), which was significantly higher than XGBoost (0.975), LR (0.774), and SVM (0.768) models (Table 4 ). Nevertheless, the values for all models were higher than 0.7, which pretends that all four prediction models have good prediction potential in the training environment. Contrarily, the models once transferred to the test set showed obvious differentiation in their performances: LR became the best-performing model with an AUC of 0.768 (95% CI, 0.642–0.893), followed by SVM (0.751), XGBoost (0.725), and RF (0.686) in descending order. Among them, the performance of RF decreased significantly. Further analysis of other indicators showed that LR had an advantage in accuracy (0.784) and specificity (0.741), while XGBoost had the highest sensitivity of 0.836, but its specificity was only 0.600. Although the accuracies of SVM and RF were 0.757, but their sensitivities were above 0.7. Table 4 Four ML prediction models’ outcomes Model AUC Accuracy Sensitivity Specificity 95% CI LR 0.768 0.784 0.700 0.741 0.642,0.893 XGBoost 0.725 0.773 0.836 0.600 0.584,0.867 SVM 0.751 0.757 0.773 0.625 0.620,0.882 RF 0.686 0.757 0.810 0.563 0.533,0.839 In this study, when 177 cases in the training set were included for statistical analysis, whether the patient was diagnosed with appendiceal gangrene and perforation was used as the outcome indicator for model building. The Lasso regression algorithm was used for feature dimensionality reduction to eliminate the influence of multicollinearity among variables, and finally, the characteristic variables with significant predictive value were screened out. Figure 1 objectively demonstrates the dynamic adjustment process of the model parameters' shrinkage under different λ values. In this study, regression models of feature subsets, based on λmin and λ1se were constructed simultaneously. To moderately relax the restrictions on the model complexity and avoid overfitting the noisy data, this study selected to include the 9 features corresponding to the λ1se value within one standard error of the model error for the subsequent construction of the machine learning prediction model. The features screened by the Lasso regression include 9 features: smoking history, fever history, WBC, percentage of lymphocytes, CRP, PCT, Alb, serum Na + level, and whether the appendiceal wall is thickened. 3.5 Model interpretation and individual analysis Firstly, we conducted univariate and multivariate LR analyses on the 9 features screened out by the Lasso regression respectively (Table 5 ). The results indicated that all the 9 features showed significant differences between the gangrene and perforation group and the non-gangrene and perforation group in the univariate logistic regression analysis, and among them, 3 features were identified as independent risk factors in the multivariate logistic regression analysis (Table 5 ). The three independent risk factors are: WBC (OR = 1.12, 95% CI, 1.01–1.25, P = 0.038), CRP (OR = 1.01, 95% CI, 1.00–1.01, P = 0.041), and Alb (OR = 0.85, 95% CI, 0.76–0.95, P = 0.004). Table 5 Single and multiple factors LR results of nine characteristics Characteristics Single factors LR Multiple factors LR OR (95% CI) P-value OR (95% CI) P-vlaue Smoking 3.45 (1.41–8.47) 0.007 2.72 (0.90–8.20) 0.076 Fever 4.33 (1.93–9.71) < 0.001 2.03 (0.74–5.51) 0.167 WBC 1.16 (1.06–1.26) < 0.001 1.12 (1.01–1.25) 0.038 LYM% 0.91 (0.86–0.97) 0.002 0.95 (0.88–1.03) 0.221 PCT 1.04 (1.01–1.08) 0.007 1.01 (0.97–1.05) 0.640 CRP 1.01 (1.01–1.02) < 0.001 1.01 (1.00-1.01) 0.041 ALB 0.84 (0.76–0.92) < 0.001 0.85 (0.76–0.95) 0.004 Na+ 0.87 (0.78–0.97) 0.011 0.94 (0.85–1.03) 0.194 Appendiceal wall thickening 0.986 According to the regression coefficients of the features (Table 6 ), our LR model is calculated using the following formula: logit(Y) = -2.0562 + 0.1086*WBC + 0.0142*CRP − 0.0335*Alb (binary predictive features take values of 0 or 1 in the formula). Table 6 Three characteristics of the LR model Characteristics Coefficient Standard error Z-statistics Pr(>|Z|) Intercept -2.0562 3.1997 -0.64 0.5205 WBC 0.1086 0.0829 1.31 0.1901 PCT 0.0142 0.0060 2.38 0.0171 ALB -0.0335 0.0752 -0.45 0.6560 The XGBoost prediction framework uses the gradient - boosted decision tree ensemble architecture to achieve modeling optimization. Its core mechanism is to build weak learners iteratively in stages. During each tree-structure splitting process, the information gain index for splitting nodes was calculated based on second-order Taylor expansion of the loss function, which is used to quantify the marginal improvement effect of candidate features on the model performance. By recursively accumulating the splitting gain values of each feature in all decision trees, the system generates global weighted statistics of multi-dimensional predictive variables and finally constructs a hierarchical feature importance evaluation system. To interpret the effect of specified features, we applied SHAP and the summary plot showed that CRP has the highest prediction importance among clinical features. Its SHAP value is significantly higher than that of other features, indicating that CRP has the greatest global impact on the model output, while the other laboratory indicators such as Alb and WBC follow closely. During the training stage of the XGBoost prediction model, its AUC reached 0.975 (95% CI, 0.953–0.998), with an accuracy rate of 93.8%. Meanwhile, the sensitivity and specificity were maintained at 93.9%, and 93.2%, respectively. While, in the test set, the AUC was 0.725 (95% CI, 0.584–0.867), and the overall prediction accuracy rate dropped to 77.3%. Meanwhile, the sensitivity index remained at 83.6%, but the specificity decreased significantly to 60.0%. The method of SVM algorithm involves introducing features with different ranks one by one with CRP, WBC, and Alb in sequence. The choice of kernel function has a decisive impact on the performance of the SVM model. By comparing the AUCs of four linear kernel functions-Linear, Polynomial, Radial, and Sigmoid-on the test set, were 0.751, 0.671, 0.663, and 0.588 respectively. Based on the linear kernel function with the highest AUC value (Linear Kernel), we constructed the SVM model and obtained a three-dimensional decision boundary graph of the training set (Fig. 3 ). During the training stage the AUC of the SVM prediction model reached 0.768 (95%Cl, 0.684–0.852), with 76.8% accuracy and 77% sensitivity with 75% specificity. In the test set, the model maintained an AUC of 0.751 (95%Cl, 0.620–0.882), with slightly decreased accuracy rate (75.7%). Meanwhile, the sensitivity increased to 77.3%, but the specificity decreased to 62.5%. Next, we evaluated the contribution intensity of features to optimize the decision path during the node splitting process (Fig. 4 ). The MDA analysis showed the ranking as WBC (13.306), CRP (8.880), and Alb (4.430). Similar trend was observed in the MDG analysis as WBC (24.255), CRP (23.788), and Alb (21.868) in sequence. We further selected 13 decision trees with the lowest classification error rate to construct the RF model. During the training stage, the model shows an AUC of up to 0.999 (95%Cl, 0.998–1.000), with an accuracy rate of 99.4%, and maintained a sensitivity and specificity of 99.2% and 100%, respectively. While, in the test set, the model's performance undergoes a significant drop in AUC (0.686 (95%Cl, 0.533–0.839)), accuracy (75.7%), sensitivity (81.0%), and the specificity (56.3%). The ROC curve based on the training set shows (Fig. 5 ) that there are significant differences in the AUCs of the four machine learning methods. Specifically, the RF model demonstrates excellent performance in the classification task (AUC 0.999, 95% CI, 0.998–1.000), and its prediction performance is superior to that of XGBoost (AUC 0.975, 95% CI, 0.953–0.998), LR (AUC 0.774, 95% CI, 0.692–0.856), and SVM (AUC 0.768, 95% CI, 0.684–0.852). The AUCs of all models exceeded 0.700, indicating that the prediction performances of these models on the training set are good. In the test dataset, the ROC curves of the four prediction models are shown in Fig. 6 . Among them, the LR model performs the best with an AUC value of 0.768 (95% CI, 0.642–0.893) as compared to SVM is 0.751 (95% CI, 0.620–0.882), XGBoost is 0.725 (95% CI, 0.584–0.867), and RF is 0.686 (95% CI, 0.533–0.839), pretending RF model exhibits more significant characteristics of performance attenuation. 3.6 Calibration analysis of LR model In this study, the LR model was finally determined as the optimal prediction tool. The verification results showed that the model demonstrated excellent comprehensive performance on the test set, with an AUC of 0.768 (95% CI, 0.642–0.893), sensitivity and specificity of 0.700, and 0.741, respectively of which were significantly better than other models. To further evaluate the reliability of the model, we used the Bootstrap resampling method (1000 repetitions) for internal calibration verification (Fig. 7 ). The results showed that the calibration curves of the training set and the test set were highly consistent with the ideal reference line (y = x), with an average absolute error of 0.023, and there was no statistical difference in the Hosmer-Lem show test (P = 0.67), confirming that there was a good consistency between the model's predicted probability and the actual occurrence probability. 3.7 Clinical applicability analysis of LR model Next, we systematically evaluated the clinical application value of the LR model in elderly patients with gangrenous perforation of acute appendicitis by plotting the DCA curve of the LR model (Fig. 8 ). Results showed that the DCA curve of the LR model is always higher than the reference lines of the extreme strategies of "full intervention" and "no intervention". This characteristic of the curve distribution indicates that this model is used to guide clinical decision-making by generating a significant net clinical benefits within a wide range of risk probability intervals. 3.8 Nomogram of LR model The LR model was then transformed into a visual clinical tool by constructing a Nomogram (Fig. 9 ), which integrated the combined predictive effects of three biomarkers, namely WBC, CRP, and Alb. Based on the quantitative framework of the synergistic effect of multiple indicators, this tool intuitively displays the risk contribution weights of each indicator through scale line segments in a two-dimensional coordinate system. Gastrointestinal surgeons can quickly read the scores of each indicator of the patient through the horizontal axis, obtain the total score after vertical accumulation, and finally locate the corresponding probability of gangrenous perforation on the risk probability axis. 4. Discussion Acute appendicitis in the elderly tends to progress more rapidly and severely than younger individuals and this accelerated progression is attributed to the unique pathophysiology[ 29 ]. When the lumen of the appendix is obstructed, bacteria proliferate within the closed environment, triggering a severe inflammatory response which in turn impairs the blood supply to the appendix, leading to gangrene and perforation, hereby exacerbating the inflammatory cascade[ 30 ]. Clinically, this is manifested as initial periumbilical pain that migrates to the right lower quadrant, accompanied by gastrointestinal symptoms such as nausea, vomiting, and abdominal distension, as well as systemic symptoms like fever[ 31 ]. However, the disease progresses more rapidly in elderly patients due to diminished physiological functions and weakened inflammatory response. Once gangrene and perforation occur, the spread of inflammation can easily lead to severe peritonitis, extensive intra-abdominal infection, and even septic shock, severely compromising the function of multiple organ systems[ 32 ]. Given the vulnerability of elderly patients to prolonged illness, it is crucial to ensure the timely medical intervention. The diagnosis of gangrenous perforation in elderly acute appendicitis is clinically challenging due to frequently atypical symptoms, such as mild abdominal pain, minimal gastrointestinal reactions, and less pronounced fever, which can easily be ignored by both patients and clinicians[ 33 ]. Laboratory tests including WBC count and neutrophil percentage may not show significant elevation, often reflecting the blunted inflammatory response is in this population. The malposition of appendix or intra-abdominal gas often interfere the imaging modalities such as abdominal ultrasound hereby limiting the diagnosis of appendiceal gangrene and perforation. Although CT is highly valuable for diagnosis, however, the health conditions of some elderly patients may not tolerate the procedure. Furthermore, the presence of multiple comorbidities in elderly patients can mask the symptoms of appendiceal gangrene and perforation thereby increasing the likelihood of misdiagnosis or delayed diagnosis[ 34 ]. ML has emerged as a powerful tool in the development of diagnostic models for various diseases, demonstrating the superior performance compared to traditional statistical methods[ 35 ]. Unlike conventional approaches, ML enables computers to "learn" from data without explicit programming, allowing the efficient handling of large datasets through complex interactions. By uncovering intricate patterns and associations within the data, ML provides the researchers with novel analytical perspectives and is increasingly playing a vital role in the medical research[ 36 ]. Our study identified WBC, CRP and Alb as independent risk factors for gangrenous perforation in elderly acute appendicitis. Utilizing ML techniques, we constructed four models—LR, RF, XGBoost, and SVM—and evaluated their performance using metrics such as AUC, sensitivity, and specificity. In the training dataset, the RF model exhibited excellent discriminative ability with an AUC of 0.999, significantly outperforming the XGBoost model (AUC = 0.975) and the LR model (AUC = 0.774). The superior performance of the RF model may be attributed to its ensemble of multiple decision trees, which allows it to capture complex interactions among features. However, this high degree of fit also suggests a potential risk of overfitting. In the test dataset, the LR model achieved the highest AUC (0.768), followed by the SVM model (0.751) and the XGBoost model (0.725), while the RF model's AUC dropped significantly to 0.686, indicating weaker generalizability. In terms of sensitivity and specificity, the XGBoost model had the highest sensitivity (0.836) but lower specificity (0.600), which may lead to a higher false-positive rate. In contrast, the LR model achieved a better balance between sensitivity (0.700) and specificity (0.741), making it more suitable for clinical decision-making that requires balanced control of false negatives and false positives. The LR model's DCA demonstrated net benefit within a threshold probability range of 10 to 50%, suggesting its potential utility in clinical decision-making. The LR model was further translated into a clinical tool through the development of a nomogram, which integrates the combined predictive effects of WBC, CRP, and Alb. This visual tool allows clinicians to quickly assess the risk of gangrenous perforation by summing the scores of individual biomarkers and locating the corresponding probability on the risk axis. The nomogram provides a quantitative framework for risk assessment based on the synergistic effects of multiple indicators, facilitating rapid and accurate decision-making. The LR model developed in this study is particularly suitable for rapid decision-making in regions with limited medical resources. The three biomarkers (WBC, CRP, and Alb) are widely used in primary care settings, eliminating the need for advanced imaging techniques such as CT or MRI[ 37 ]. This approach also reduces diagnostic costs significantly. However, the study's limitations include its single-center retrospective design and relatively small sample size, which may affect the model's generalizability. Future work should focus on validating the model through multicenter, prospective cohort studies. Additionally, efforts should be directed towards establishing a standardized database across different regions, incorporating elderly patients with diverse demographic and comorbid profiles, to enhance the model's adaptability to population heterogeneity. Conclusion This study demonstrates that the developed machine learning model markedly enhances the diagnostic accuracy for gangrenous perforation in acute appendicitis among elderly patients, outperforming conventional analytical approaches. From an initial set of 38 clinical indicators, three key predictors—WBC, CRP, and Alb—were identified as critical for model development. In the RF model, WBC emerged as the most influential feature, whereas CRP was the dominant predictor in the other three models evaluated. Among the four models assessed, the LR model achieved the optimal balance of AUC, sensitivity, and specificity, despite the XGBoost model exhibiting the highest sensitivity. Given its interpretability and practical applicability in clinical settings, the LR model is considered the most effective. Furthermore, the LR model's nomogram offers an intuitive, visually accessible tool, enabling gastrointestinal surgeons to make swift and informed diagnostic decisions for elderly patients with gangrenous perforation of acute appendicitis. Declarations Ethics approval and consent to participate This study was approved by the Second Affiliated Hospital of Kunming Medical University (Approval No. PJ-2024-185). All methods were conducted in accordance with relevant guidelines and regulations, including the Declaration of Helsinki. Informed consent was not required due to the retrospective design of the study and the exemption granted by the IRB. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Funding This study was supported by the Science and Technology Plan Project of the Department of Science and Technology of Yunnan Province (202501AY070001-030) and Yunnan Revitalization Talent Support Program (XDYC-MY-2022-0042 & YNWR-MY-2020-050). Author Contribution S.F conceived the study. C.J, S.Y.B and Y.B designed the study. Y.B, L.Y.L and S.Y.B discovered the clinical effect. C.J, S.Y.Q.L and F.D analyzed the data. C.J and G.Y wrote the initial draft. S.F revised the manuscript. All authors read and approved the final version of manuscript for publication. Acknowledgements Not applicable. Data Availability The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request. References Powell SK. Caring for Caregivers: National Case Management Week: October 13–10, 2024. Volume 29. LWW; 2024. pp. 187–8. Sisik A, Kudas I, Basak F, Hasbahceci M. Is the increased incidence of pathologically proven acute appendicitis more likely seen in elderly patients? A retrospective cohort study. Aging Male. 2021;24(1):1–7. Khan HT. Population ageing in a globalized world: Risks and dilemmas? J Eval Clin Pract. 2019;25(5):754–60. Dhillon NK, Barmparas G, Lin T-L, Alban RF, Melo N, Yang AR, Margulies DR, Ley EJ. Unexpected complicated appendicitis in the elderly diagnosed with acute appendicitis. Am J Surg. 2019;218(6):1219–22. Segev L, Keidar A, Schrier I, Rayman S, Wasserberg N, Sadot E. Acute appendicitis in the elderly in the twenty-first century. J Gastrointest Surg. 2015;19(4):730–5. Yuan J, Chen Q, Hong W, Yu L, Li X. Comparison of clinical features and outcomes of appendectomy in elderly vs. non-elderly: A systematic review and meta-analysis. Front Surg. 2022;9:818347. Arlt A, Bharti R, Ilves I, Häsler R, Miettinen P, Paajanen H, Brunke G, Ellrichmann M, Rehman A, Hauser C. Characteristic changes in microbial community composition and expression of innate immune genes in acute appendicitis. Innate Immun. 2015;21(1):30–41. Jung SK, Rhee DY, Lee WJ, Woo SH, Seol SH, Kim DH, Choi SP. Neutrophil-to-lymphocyte count ratio is associated with perforated appendicitis in elderly patients of emergency department. Aging Clin Exp Res. 2017;29:529–36. Omari AH, Khammash MR, Qasaimeh GR, Shammari AK, Yaseen MKB, Hammori SK. Acute appendicitis in the elderly: risk factors for perforation. World J Emerg Surg. 2014;9:1–6. Lapsa S, Ozolins A, Strumfa I, Gardovskis J. Acute appendicitis in the elderly: a literature review on an increasingly frequent surgical problem. Geriatrics. 2021;6(3):93. Frenzel F, Hollaender S, Fries P, Stroeder R, Stroeder J. Jejunal obstruction due to rare internal hernia between skeletonized external iliac artery and vein as late complication of laparoscopic hysterectomy with pelvic lymphadenectomy—case report and review of literature. Arch Gynecol Obstet. 2020;302:1075–80. Cimino MM, Biloslavo A, Kurihara H, Bellio G, Porta M, Fattori S, Bass GA. Practice patterns and clinical outcomes in acute appendicitis differ in the elderly patient. Eur J Trauma Emerg Surg 2024:1–10. Al-Gaithy ZK. Clinical value of total white blood cells and neutrophil counts in patients with suspected appendicitis: retrospective study. World J Emerg Surg. 2012;7:1–7. Kim JJ, Dobson BH, Ng LH, Thong C, Arthur DW, Parker T, Collaboration D, Anwari Q, Archer T, Auld L. Can normal inflammatory markers rule out acute appendicitis? The reliability of biochemical investigations in diagnosis. ANZ J Surg. 2020;90(10):1970–4. Hamid MA, Afroz R, Ahmed UN, Bawani A, Khan D, Shahab R, Salim A. The importance of visualization of appendix on abdominal ultrasound for the diagnosis of appendicitis in children: A quality assessment review. World J Emerg Med. 2020;11(3):140. Perrotta G, Geropoulos G, Bhan C. The role of imaging in the diagnosis of acute appendicitis during the COVID-19 pandemic: a retrospective cohort study. Updates Surg. 2023;75(1):205–8. Tulin-Silver S, Babb J, Pinkney L, Strubel N, Lala S, Milla SS, Tomita S, Fefferman NR. The challenging ultrasound diagnosis of perforated appendicitis in children: constellations of sonographic findings improve specificity. Pediatr Radiol. 2015;45:820–30. Park JH, Salminen P, Tannaphai P, Lee KH. Low-dose abdominal CT for evaluating suspected appendicitis in adolescents and young adults: review of evidence. Korean J Radiol. 2022;23(5):517. Kim D, Woodham BL, Chen K, Kuganathan V, Edye MB. Rapid MRI abdomen for assessment of clinically suspected acute appendicitis in the general adult population: a systematic review. J Gastrointest Surg. 2023;27(7):1473–85. Radak M, Lafta HY, Fallahi H. Machine learning and deep learning techniques for breast cancer diagnosis and classification: a comprehensive review of medical imaging studies. J Cancer Res Clin Oncol. 2023;149(12):10473–91. Goecks J, Jalili V, Heiser LM, Gray JW. How machine learning will transform biomedicine. Cell. 2020;181(1):92–101. Salvador–Meneses J, Ruiz–Chavez Z, Garcia–Rodriguez J. Compressed k NN: K-nearest neighbors with data compression. Entropy. 2019;21(3):234. Breiman L. Random forests. Mach Learn. 2001;45:5–32. Hou N, Li M, He L, Xie B, Wang L, Zhang R, Yu Y, Sun X, Pan Z, Wang K. Predicting 30-days mortality for MIMIC-III patients with sepsis-3: a machine learning approach using XGboost. J translational Med. 2020;18:1–14. Stoltzfus JC. Logistic regression: a brief primer. Acad Emerg Med. 2011;18(10):1099–104. Zhang X, Song Q. A multi-label learning based kernel automatic recommendation method for support vector machine. PLoS ONE. 2015;10(4):e0120455. Jiang H, Ching W-K. Correlation kernels for support vector machines classification with applications in cancer data. Comput Math Methods Med. 2012;2012(1):205025. Seshan VE, Gönen M, Begg CB. Comparing ROC curves derived from regression models. Stat Med. 2013;32(9):1483–93. Cohen-Arazi O, Dabour K, Bala M, Haran A, Almogy G. Management, treatment and outcomes of acute appendicitis in an elderly population: a single-center experience. Eur J Trauma Emerg Surg. 2017;43(5):723–7. Tran N, Emuakhagbon VS, Baker BT, Huerta S. Re-assessing the role of the fecalith in acute appendicitis in adults: case report, case series and literature review. J Surg Case Rep. 2021;2021(1):rjaa543. Soldo I, Radisic Biljak V, Bakula B, Bakula M, Simundic AM. The diagnostic accuracy of clinical and laboratory parameters in the diagnosis of acute appendicitis in the adult emergency department population - a case control pilot study. Biochem Med (Zagreb). 2018;28(3):030712. Ikeda K, Ikeda T, Tani N, Shida A, Aoki Y, Morioka F, Ishikawa T. Acute abdominal disease with nonspecific symptoms in the elderly: A case series. J Forensic Leg Med. 2021;78:102089. Fugazzola P, Ceresoli M, Agnoletti V, Agresta F, Amato B, Carcoforo P, Catena F, Chiara O, Chiarugi M, Cobianchi L, et al. The SIFIPAC/WSES/SICG/SIMEU guidelines for diagnosis and treatment of acute appendicitis in the elderly (2019 edition). World J Emerg Surg. 2020;15(1):19. Siddharthan RV, Byrne RM, Dewey E, Martindale RG, Gilbert EW, Tsikitis VL. Appendiceal cancer masked as inflammatory appendicitis in the elderly, not an uncommon presentation (Surveillance Epidemiology and End Results (SEER)-Medicare Analysis). J Surg Oncol. 2019;120(4):736–9. Bi Q, Goodman KE, Kaminsky J, Lessler J. What is Machine Learning? A Primer for the Epidemiologist. Am J Epidemiol. 2019;188(12):2222–39. Beunza JJ, Puertas E, García-Ovejero E, Villalba G, Condes E, Koleva G, Hurtado C, Landecho MF. Comparison of machine learning algorithms for clinical event prediction (risk of coronary heart disease). J Biomed Inf. 2019;97:103257. Oymaci E, Kahramansoy N, Tan S, Aydogan S, Yildirim M. The diagnostic role of preoperative blood tests in complicated appendicitis: A feasible approach to surgical decision. Niger J Clin Pract. 2023;26(7):1005–10. Additional Declarations No competing interests reported. Supplementary Files SupplementaryMaterial.docx Cite Share Download PDF Status: Published Journal Publication published 24 Apr, 2026 Read the published version in BMC Surgery → Version 1 posted Editorial decision: Revision requested 27 Feb, 2026 Reviews received at journal 27 Feb, 2026 Reviewers agreed at journal 27 Feb, 2026 Reviewers agreed at journal 26 Feb, 2026 Reviews received at journal 18 Feb, 2026 Reviews received at journal 10 Feb, 2026 Reviewers agreed at journal 06 Feb, 2026 Reviewers agreed at journal 06 Feb, 2026 Reviewers invited by journal 13 Oct, 2025 Editor invited by journal 29 Sep, 2025 Editor assigned by journal 26 Sep, 2025 Submission checks completed at journal 26 Sep, 2025 First submitted to journal 19 Sep, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7656285","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":534708327,"identity":"f0a8e6b2-dcac-4f64-83e7-d0d6c0f6de96","order_by":0,"name":"Jie Cheng","email":"","orcid":"","institution":"The Second Affiliated Hospital of Kunming Medical University","correspondingAuthor":false,"prefix":"","firstName":"Jie","middleName":"","lastName":"Cheng","suffix":""},{"id":534708328,"identity":"90129f1f-286f-4ce5-a07d-90f2eff1d6bf","order_by":1,"name":"Yanbo Sun","email":"","orcid":"","institution":"The Second Affiliated Hospital of Kunming Medical University","correspondingAuthor":false,"prefix":"","firstName":"Yanbo","middleName":"","lastName":"Sun","suffix":""},{"id":534708329,"identity":"3ff004f8-25e7-4297-972a-505b9bd3ba3c","order_by":2,"name":"Bin Yang","email":"","orcid":"","institution":"The Second Affiliated Hospital of Kunming Medical University","correspondingAuthor":false,"prefix":"","firstName":"Bin","middleName":"","lastName":"Yang","suffix":""},{"id":534708330,"identity":"a166271f-88fb-4efd-80df-00f29e916a88","order_by":3,"name":"Yun Gong","email":"","orcid":"","institution":"The Second Affiliated Hospital of Kunming Medical University","correspondingAuthor":false,"prefix":"","firstName":"Yun","middleName":"","lastName":"Gong","suffix":""},{"id":534708331,"identity":"d9a4f154-ae87-419b-90d2-21dddb5ddd6e","order_by":4,"name":"Die Fan","email":"","orcid":"","institution":"The Third Affiliated Hospital of Kunming Medical University","correspondingAuthor":false,"prefix":"","firstName":"Die","middleName":"","lastName":"Fan","suffix":""},{"id":534708332,"identity":"8b641d53-8f77-44c9-9221-c08950439530","order_by":5,"name":"Yongqilin Shuang","email":"","orcid":"","institution":"Diqing Tibetan Autonomous Prefectural People's Hospital","correspondingAuthor":false,"prefix":"","firstName":"Yongqilin","middleName":"","lastName":"Shuang","suffix":""},{"id":534708333,"identity":"69c9beba-e926-4e9b-81a1-0d5906c2ed60","order_by":6,"name":"Yanli Li","email":"","orcid":"","institution":"Yan’an Hospital of Kunming City","correspondingAuthor":false,"prefix":"","firstName":"Yanli","middleName":"","lastName":"Li","suffix":""},{"id":534708334,"identity":"eaf44578-67c5-4a0b-984c-906c1e97b867","order_by":7,"name":"Feng Sun","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA5ElEQVRIiWNgGAWjYJCCD0DMwy//sPFBQkUNUToYZwAJGcmG5GaDB2eOEa/FxuBAepvkwxZmwurlZ+QebPi4o5bH4MDBtorEBjYG/vbuBLxaDG7kJTbOPHOcR/JgY9uNxB0yDBJnzm7Ar0Uix/wxb9sxHr7DjEAtZ9iAIrn4tcjPyDFs/gvUwnCMsa0gsY2ZsBaGG0AtjG01PAJnGNsYiNJicOaNYWNv2wEeyRmMzRIJZ47xEPSLfHuOYcPPtjp7fgn2hx9/VNTI8bf3EnAYBByGs3iIUQ4CdcQqHAWjYBSMgpEIAKZHTpF1TpskAAAAAElFTkSuQmCC","orcid":"","institution":"The Second Affiliated Hospital of Kunming Medical University","correspondingAuthor":true,"prefix":"","firstName":"Feng","middleName":"","lastName":"Sun","suffix":""}],"badges":[],"createdAt":"2025-09-19 08:38:17","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7656285/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7656285/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s12893-026-03753-y","type":"published","date":"2026-04-24T15:58:41+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":94622870,"identity":"abc855ef-0206-4768-bc73-8020c29ebf4e","added_by":"auto","created_at":"2025-10-29 04:18:37","extension":"jpg","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":56852,"visible":true,"origin":"","legend":"","description":"","filename":"Figer1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/9674c2e827aa7a0996502758.jpg"},{"id":94622854,"identity":"19bd97bd-3f14-4f12-8c9a-a22d326f898a","added_by":"auto","created_at":"2025-10-29 04:18:36","extension":"jpg","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":54899,"visible":true,"origin":"","legend":"","description":"","filename":"Figer2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/e60b1b0de88b96d386ceba99.jpg"},{"id":94622613,"identity":"9bc67dfd-8d21-4db6-b38e-7d0cdf9a3bd8","added_by":"auto","created_at":"2025-10-29 04:18:24","extension":"jpg","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":56770,"visible":true,"origin":"","legend":"","description":"","filename":"Figer3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/126340ca7c76a2a87fe76715.jpg"},{"id":94622639,"identity":"23c6fc16-3371-48af-bd2e-f08bb9a1f36d","added_by":"auto","created_at":"2025-10-29 04:18:26","extension":"docx","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":71072,"visible":true,"origin":"","legend":"","description":"","filename":"Manuscript.docx","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/82b1c819c947f987287aa910.docx"},{"id":94622239,"identity":"a6dd0495-25e3-4451-8eb8-dd0d3ee3902a","added_by":"auto","created_at":"2025-10-29 04:18:12","extension":"docx","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":33744,"visible":true,"origin":"","legend":"","description":"","filename":"Tables.docx","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/df2b8c983cd915c20168fde9.docx"},{"id":94622369,"identity":"17fa842d-0699-4e27-8d82-37634eb5a92f","added_by":"auto","created_at":"2025-10-29 04:18:17","extension":"jpg","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":25393,"visible":true,"origin":"","legend":"","description":"","filename":"Figer4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/ccae0aa457d8a7d9223dd344.jpg"},{"id":94622948,"identity":"f7719d25-9718-466f-9525-f938319e46b5","added_by":"auto","created_at":"2025-10-29 04:18:40","extension":"jpg","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":48548,"visible":true,"origin":"","legend":"","description":"","filename":"Figer5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/de7a96bb9ab75d0169f07740.jpg"},{"id":94622505,"identity":"fc22cfe4-acd0-43bc-af90-5f44f39527be","added_by":"auto","created_at":"2025-10-29 04:18:21","extension":"jpg","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":51848,"visible":true,"origin":"","legend":"","description":"","filename":"Figer6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/033d22b37273be8ba667ae37.jpg"},{"id":94622545,"identity":"58532de9-30b8-4335-807b-fc5e11adc156","added_by":"auto","created_at":"2025-10-29 04:18:22","extension":"jpg","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":68755,"visible":true,"origin":"","legend":"","description":"","filename":"Figer7.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/e6237a04af61997cc7e08117.jpg"},{"id":94622182,"identity":"7f14791a-2e35-4191-9011-0a24a632d2ef","added_by":"auto","created_at":"2025-10-29 04:18:08","extension":"jpg","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":43955,"visible":true,"origin":"","legend":"","description":"","filename":"Figer8.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/64cd676fb3952259ea4ea96f.jpg"},{"id":94622724,"identity":"8735206f-6af4-421d-b041-3cb9bce61db0","added_by":"auto","created_at":"2025-10-29 04:18:29","extension":"jpg","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":153736,"visible":true,"origin":"","legend":"","description":"","filename":"Figer9.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/c7618165c43c5dcd60ee769e.jpg"},{"id":94622656,"identity":"ae3ee977-a466-4549-8fe2-a8badf0c57f3","added_by":"auto","created_at":"2025-10-29 04:18:26","extension":"json","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":9364,"visible":true,"origin":"","legend":"","description":"","filename":"b448882ff5be46daaaab515574b2d2dd.json","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/8122071996c6951ec2c203ad.json"},{"id":94622934,"identity":"46021a2f-9d06-4198-8ce9-8eb6cc0b7baf","added_by":"auto","created_at":"2025-10-29 04:18:39","extension":"docx","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":170253,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryMaterial.docx","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/fc1549dc351147c59d903703.docx"},{"id":94622835,"identity":"52868bab-5c05-46a0-a73e-0aab37ab313a","added_by":"auto","created_at":"2025-10-29 04:18:34","extension":"xml","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":164246,"visible":true,"origin":"","legend":"","description":"","filename":"b448882ff5be46daaaab515574b2d2dd1enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/d370ded6efc3fe4e16830da7.xml"},{"id":94622682,"identity":"f7102ad1-c9a5-4831-a998-d7182a08e9a3","added_by":"auto","created_at":"2025-10-29 04:18:27","extension":"jpg","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":56852,"visible":true,"origin":"","legend":"","description":"","filename":"Figer1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/db4ea7be305095f3afe968c0.jpg"},{"id":94622665,"identity":"2cef2b60-5b2b-4525-84b6-26a640637cb2","added_by":"auto","created_at":"2025-10-29 04:18:26","extension":"jpg","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":54899,"visible":true,"origin":"","legend":"","description":"","filename":"Figer2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/bfe7dd03b6952d2b14325a5d.jpg"},{"id":94622244,"identity":"5f16f793-0304-4b9b-b9f1-aaccd21db4b4","added_by":"auto","created_at":"2025-10-29 04:18:13","extension":"jpg","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":56770,"visible":true,"origin":"","legend":"","description":"","filename":"Figer3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/b732c13616e49b2b900dcfe6.jpg"},{"id":94622207,"identity":"f01a2cc1-d486-4ba7-a837-3d8b0a737247","added_by":"auto","created_at":"2025-10-29 04:18:11","extension":"jpg","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":25393,"visible":true,"origin":"","legend":"","description":"","filename":"Figer4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/d2c4e32023533d6af84ebb6e.jpg"},{"id":94622942,"identity":"6a0f0616-973b-4441-8e7f-a0572cfc4b6b","added_by":"auto","created_at":"2025-10-29 04:18:39","extension":"jpg","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":48548,"visible":true,"origin":"","legend":"","description":"","filename":"Figer5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/6e15b88eec0d338535d8e85e.jpg"},{"id":94622368,"identity":"7a75ecc5-8dbb-481d-816f-332e05572c0e","added_by":"auto","created_at":"2025-10-29 04:18:17","extension":"jpg","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":51848,"visible":true,"origin":"","legend":"","description":"","filename":"Figer6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/e8ac6fb8f74dd86360005229.jpg"},{"id":94622664,"identity":"aa4e8013-6068-4d4e-8535-8e3a32706d16","added_by":"auto","created_at":"2025-10-29 04:18:26","extension":"jpg","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":68755,"visible":true,"origin":"","legend":"","description":"","filename":"Figer7.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/760c25ab76202af7fb9d437a.jpg"},{"id":94622739,"identity":"29192c23-bfc2-41c0-bc77-63c19748d284","added_by":"auto","created_at":"2025-10-29 04:18:31","extension":"jpg","order_by":21,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":43955,"visible":true,"origin":"","legend":"","description":"","filename":"Figer8.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/2801b1ef5b472d6467f705a7.jpg"},{"id":94622617,"identity":"79aeb82e-21ed-4dc8-a859-1653ed7332c7","added_by":"auto","created_at":"2025-10-29 04:18:24","extension":"jpg","order_by":22,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":153736,"visible":true,"origin":"","legend":"","description":"","filename":"Figer9.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/200fbe4e2b0af50a88dd293f.jpg"},{"id":94622311,"identity":"7907e82d-2642-4692-a722-6ee8ba64f949","added_by":"auto","created_at":"2025-10-29 04:18:16","extension":"wmf","order_by":23,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":484,"visible":true,"origin":"","legend":"","description":"","filename":"image1.wmf","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/b12ff8c4033aefcfad581402.wmf"},{"id":94623011,"identity":"2bd24b2b-c0a5-46a8-bcdd-fc94a1abb9f2","added_by":"auto","created_at":"2025-10-29 04:18:43","extension":"png","order_by":24,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":31819,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFiger1.png","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/5ba1b75b4d87a052062954e6.png"},{"id":94622271,"identity":"a89fcfee-95fc-48ee-b295-c53439aacb71","added_by":"auto","created_at":"2025-10-29 04:18:14","extension":"png","order_by":25,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":24694,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFiger2.png","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/b94491c7d0ad69bd6994c80d.png"},{"id":94622956,"identity":"0b0ef197-b554-492b-8e27-4ce261f6f138","added_by":"auto","created_at":"2025-10-29 04:18:40","extension":"png","order_by":26,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":33018,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFiger3.png","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/e92bdb2b1cad58ffc155d465.png"},{"id":94622309,"identity":"b6741ddb-518c-447e-bb8d-fbf67de8b9d8","added_by":"auto","created_at":"2025-10-29 04:18:16","extension":"png","order_by":27,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":18676,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFiger4.png","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/23cca3049af98ce99e70e184.png"},{"id":94622282,"identity":"484f2bbc-a589-4fb5-bf86-1e982321cff1","added_by":"auto","created_at":"2025-10-29 04:18:15","extension":"png","order_by":28,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":25579,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFiger5.png","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/a5f8d63e1a20a4a3a43d0d98.png"},{"id":94622362,"identity":"709c7108-aa6d-4f93-bceb-f3d79c2d23fd","added_by":"auto","created_at":"2025-10-29 04:18:17","extension":"png","order_by":29,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":27981,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFiger6.png","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/a3e823214c0d0d7dd656f8e7.png"},{"id":94622844,"identity":"3d6f5872-d9cb-4e4a-9d9d-18609b9cc66a","added_by":"auto","created_at":"2025-10-29 04:18:35","extension":"png","order_by":30,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":13994,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFiger7.png","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/0b116bbe43263266faada84e.png"},{"id":94622674,"identity":"ce9085b6-675a-4d88-9b98-ff12e3a018a2","added_by":"auto","created_at":"2025-10-29 04:18:27","extension":"png","order_by":31,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":22526,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFiger8.png","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/37a8c2c73ba76d381567122b.png"},{"id":94622875,"identity":"8dc944d9-5d81-4f88-b8d6-b0de3c0f6d71","added_by":"auto","created_at":"2025-10-29 04:18:37","extension":"png","order_by":32,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":33724,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFiger9.png","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/9c9a559c295a2a97d5c55855.png"},{"id":94622365,"identity":"b48e4e47-5bae-483f-a195-9433d57f1930","added_by":"auto","created_at":"2025-10-29 04:18:17","extension":"png","order_by":33,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":350,"visible":true,"origin":"","legend":"","description":"","filename":"Onlineimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/e13c0c7c81789b669d943cb4.png"},{"id":94622978,"identity":"c013ea4b-fea2-4793-a216-fb67e1afc217","added_by":"auto","created_at":"2025-10-29 04:18:41","extension":"xml","order_by":34,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":162071,"visible":true,"origin":"","legend":"","description":"","filename":"b448882ff5be46daaaab515574b2d2dd1structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/2228d32a168521d2b758818e.xml"},{"id":94622908,"identity":"6ca4bd00-251b-4129-9d43-8720d031d82c","added_by":"auto","created_at":"2025-10-29 04:18:38","extension":"html","order_by":35,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":172338,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/3944a49cad296c98ada26cf3.html"},{"id":94622265,"identity":"195ace2c-6354-4d78-a690-cc6be31fd089","added_by":"auto","created_at":"2025-10-29 04:18:14","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":56852,"visible":true,"origin":"","legend":"\u003cp\u003eLasso regression coefficient and log (λ) value.\u003c/p\u003e","description":"","filename":"Figer1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/d1b3736efa3f66b1470b0c8b.jpg"},{"id":94622876,"identity":"6538e25c-6c04-4981-baae-ffdeaa336307","added_by":"auto","created_at":"2025-10-29 04:18:37","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":54899,"visible":true,"origin":"","legend":"\u003cp\u003eLog (λ) value and model error: the two vertical dashed lines correspond to two optimized choices of the regularization parameter λ: the λmin value corresponding to the lowest point of the cross-validation error and the λ1se threshold within one standard deviation of it. The visual analysis shows that as the logarithmic value (abscissa) of the regularization parameter λ increases, the model degrees of freedom (ordinate) show an exponential decay trend, and there is a significant negative correlation between the model complexity and the generalization ability.\u003c/p\u003e","description":"","filename":"Figer2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/5a2ec53f754e328649b3f623.jpg"},{"id":94622633,"identity":"0a22e3e3-5b30-460b-b546-f87b52d4301a","added_by":"auto","created_at":"2025-10-29 04:18:25","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":56770,"visible":true,"origin":"","legend":"\u003cp\u003eThree-dimensional Decision Boundary of the SVM Training Set.\u003c/p\u003e","description":"","filename":"Figer3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/f168e4076c04e9cd1d85c945.jpg"},{"id":94622883,"identity":"5efb082f-49f4-4b6f-8601-9a54f0a2c0bf","added_by":"auto","created_at":"2025-10-29 04:18:37","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":25393,"visible":true,"origin":"","legend":"\u003cp\u003eFeature Importance Ranking of the RF Model.\u003c/p\u003e","description":"","filename":"Figer4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/071544db411dc3aff560c491.jpg"},{"id":94622261,"identity":"a5af851d-c439-4468-b4ab-3efd6f6bee3b","added_by":"auto","created_at":"2025-10-29 04:18:14","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":48548,"visible":true,"origin":"","legend":"\u003cp\u003eROC Curves of the Training Set for Four Machine Learning Prediction Models.\u003c/p\u003e","description":"","filename":"Figer5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/ea331c36df584a3f75bc9a09.jpg"},{"id":94622897,"identity":"eaba01d4-5677-4f9c-9fd4-f8bb2937ee42","added_by":"auto","created_at":"2025-10-29 04:18:38","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":51848,"visible":true,"origin":"","legend":"\u003cp\u003eROC Curves of the Test Set for Four Machine Learning Prediction Models.\u003c/p\u003e","description":"","filename":"Figer6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/983031dff026516c0963eb21.jpg"},{"id":94622574,"identity":"871293a4-f5be-46d1-b56e-86343f715ed9","added_by":"auto","created_at":"2025-10-29 04:18:23","extension":"jpg","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":68755,"visible":true,"origin":"","legend":"\u003cp\u003eCalibration Curve of the LR Model on the \u003cstrong\u003e(A)\u003c/strong\u003eTraining Set and \u003cstrong\u003e(B)\u003c/strong\u003e Test Set.\u003c/p\u003e","description":"","filename":"Figer7.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/6e6fddd8eadb95ad5b45ee34.jpg"},{"id":94622550,"identity":"b0fc2762-0efb-43ff-8cb5-718e74f48e5a","added_by":"auto","created_at":"2025-10-29 04:18:22","extension":"jpg","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":43955,"visible":true,"origin":"","legend":"\u003cp\u003eDCA curve of the LR model on the testing set.\u003c/p\u003e","description":"","filename":"Figer8.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/4fc4c01442a889d30c93a967.jpg"},{"id":94622857,"identity":"0750aa45-6278-4d50-91a6-2801a3d2b1b4","added_by":"auto","created_at":"2025-10-29 04:18:36","extension":"jpg","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":153736,"visible":true,"origin":"","legend":"\u003cp\u003eNomogram of LR model.\u003c/p\u003e","description":"","filename":"Figer9.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/4ab9913b82956aea6e582fe5.jpg"},{"id":107928036,"identity":"8706dd8d-7416-4a39-b465-4e0e6c093301","added_by":"auto","created_at":"2026-04-27 16:06:42","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1140852,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/d903f760-a0e7-48f1-bf34-52ed5456f452.pdf"},{"id":94622868,"identity":"388586df-b78b-4b26-aec6-ace31d7eadf8","added_by":"auto","created_at":"2025-10-29 04:18:36","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":170253,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryMaterial.docx","url":"https://assets-eu.researchsquare.com/files/rs-7656285/v1/ba2386fd7c57394a2d2c8e49.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Machine learning constructs a diagnostic prediction model for gangrenous perforation of acute appendicitis in elderly patients","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eAging is advancing at an unprecedented rate all over the world as evidenced by World Health Organization that the global population of people aged 60 and above is expected to double, reaching approximately 2.1\u0026nbsp;billion by 2050, and the proportion of the population aged 60 and above will increase from 13.5% in 2020 to 22% in 2050, and this rising trend has a profound impact on the different aspects of society particularly the healthcare sector [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. The incidence rates of various geriatric diseases like acute appendicitis in the elderly patients is particularly fluctuating with a significant upward trend of 3.2% has been reported annually [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. These vigorous rising facts pretend the higher risk of acute abdominal diseases in the elderly population.\u003c/p\u003e\u003cp\u003eWith the rise in acute appendicitis in the elderly the condition also differs from young patient due to unique pathophysiological characteristics. A major concern in clinical treatment is high incidence of gangrene and perforation (30%-40%) which is 2\u0026ndash;3 times higher than young patients and poses significant challenges like as high risk of postoperative complications, prolonged hospital stay and hereby increased medical costs [\u003cspan additionalcitationids=\"CR5\" citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eThe poor prognosis in elderly patients with acute appendicitis is due to multiple factors like weak immune system, degenerative changes in blood vessels, and alteration in pain sensitivity. As immune response declines, the body's defense mechanisms become less responsive to inflammation, allowing appendiceal inflammation to progress quickly and making it difficult to control effectively. [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. The degenerative changes in blood vessels affect the blood supply of the appendix resulting in local tissue ischemia and hypoxia, thereby increasing the risk of gangrene and perforation [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. The decrease in pain sensitivity is also a factor that cannot be ignored. Similarly, the reduction in pain perception ability of elderly patients often worsen the condition due to delay in detection of pain symptoms in a timely manner at the initial stage of appendicitis [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. Thus, the clinical diagnosis and treatment become challenging in elderly population making them a high-risk group in the diagnosis and treatment of acute abdominal diseases.\u003c/p\u003e\u003cp\u003eAnother challenge is early and accurate diagnosis however, complexity of underlying diseases, decline in body functions with age, hidden or atypical clinical symptoms limits the use of traditional diagnostic methods [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. The typical migratory abdominal pain, which is a hallmark symptom of acute appendicitis, however, decrease in pain sensitivity and changes in nerve conduction often leads to absence of such typical symptoms in most elderly patients [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. Similarly, peritoneal irritation being a significant base for intra-abdominal inflammation is often not experienced by elderly patients (~\u0026thinsp;28% patients) [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. Even loose abdominal wall muscles and slow response also results in poor diagnosis or misdiagnosis [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. Irrespective of this, use to commonlaboratory tests, like as C-reactive protein (CRP) and white blood cell count (WBC) are offers a for diagnosis is limited due to inconsistent sensitivity and specificity of CRP and wide range of WBC, and also interfered with by other factors in elderly patients, resulting in the non-specific alterations in these indicators, hereby making it difficult to accurately to independently diagnose the gangrene and perforation [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eNevertheless, the imaging techniques like ultrasonography, computed tomography (CT), and magnetic resonance imaging (MRI) offers a best aid in diagnosis of acute appendicitis in the elderly patients. But intestinal gas, position of appendix, and increased adipose tissue compromise the sensitivity of non-invasive ultrasonogphy in elderly patients as compared to young patients (68% vs., 92%) [\u003cspan additionalcitationids=\"CR16\" citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. While, CT offers best alternate for ambiguous diagnosis and increased the sensitivity to 89%, for elderly patients with poor renal function but the use of contrast agents has been reported to even cause the renal failure in severe cases [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. Nonetheless, MRI being a radiation-free imaging method has shown a high accuracy in diagnosing the complex appendicitis but the cost and time and limited access hinder wide spread use in the diagnosis of appendicitis hereby necessitating the use of alternatives [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eThe rapid development of medical technology has enabled the machine learning (ML) to play an increasingly important role in the clinics, and is helpful in accurate diagnosis, and formulation and optimization of patient-based treatment plans for different diseases. Evidences from breast cancer diagnosis and classification reported that ML has achieved remarkable achievements in various medical imaging modalities such as mammography, ultrasound, MRI, histology, and thermography with high diagnostic accuracy [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]. It has improved the precision of decision-making and patients\u0026rsquo; treatment outcomes in clinics, thus offering a great potential in clinical diagnosis, personalized treatment, and health monitoring [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. Nevertheless, the application of ML technology in diagnosis of gangrene and perforation of acute appendicitis in the elderly is still in the initial exploration stage with limited reports. Thus, in this study, we utilized the ML technology to construct a diagnostic prediction model of gangrene and perforation of acute appendicitis in the elderly patients, hereby focusing on in-depth analysis of the clinical characteristics of gangrene and perforation of acute appendicitis to provide efficient and reliable auxiliary diagnostic tools for gastrointestinal surgeons.\u003c/p\u003e"},{"header":"2 Material and Methods","content":"\u003cp\u003eA retrospective study from June 2021 to January 2024 was carried out at the Second Affiliated Hospital of Kunming Medical University, China. The research protocol and consent to participate was approved by the ethical Review Committee of The Second Affiliated Hospital of Kunming Medical University, China.\u003c/p\u003e\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003e2.1 Patients\u0026rsquo; selection\u003c/h2\u003e\u003cp\u003eThe inclusion criteria for patients was as: the patients were diagnosed with acute appendicitis through a comprehensive diagnosis based on pre-operative classic clinical manifestations (migratory right lower quadrant pain, tenderness and rebound tenderness at McBurney's point), imaging features (abdominal CT showing thickening of the appendix and exudation of surrounding fat streaks), and laboratory indicators (elevated white blood cell count), and were confirmed by laparoscopic surgical resection combined with intraoperative exploration and postoperative pathological examination; having age\u0026thinsp;\u0026ge;\u0026thinsp;60 years old with complete perioperative data, including pre-operative CT imaging data, laboratory tests (complete blood count, comprehensive biochemistry, coagulation function, electrolytes), and postoperative pathological reports.\u003c/p\u003e\u003cp\u003eThe patients with following complications like active infection in other parts, history of malignant tumor or postoperative pathological examination suggesting a neoplastic lesion, abnormal coagulation function or use of anticoagulant/antiplatelet drugs within 30 days before surgery, hematological diseases, and severe hepatic and renal insufficiency were not included in the study.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\u003ch2\u003e2.2 Parameters of study\u003c/h2\u003e\u003cp\u003eThe data collected was categorized into baseline data, per-operative laboratory indicators, and imaging features and pathological and intraoperative records. Briefly the baseline data comprised of demographic characteristics (age, gender, weight, height, BMI), underlying diseases (diabetes, hypertension), lifestyle habits (smoking, drinking), symptomatic characteristics (migratory right lower quadrant pain, nausea and vomiting, anorexia, fever, tenderness and rebound tenderness in the right lower quadrant), and disease course-related indicators (time from onset to surgery, use of preoperative antibacterial drugs). Thepre-operative laboratory indicators included inflammatory markers (white blood cell count WBC, C-reactive protein CRP, procalcitonin PCT), hematological parameters (platelet PLT, hemoglobin HGB, hematocrit HCT, red blood cell distribution width RDW), metabolic indicators (albumin Alb, serum Na+, direct/indirect bilirubin DBIL/IBIL), and coagulation function (D-D dimer), and the pre-operative imaging features were the minimum diameter of the appendix, thickening of the appendiceal wall (defined as thickening when \u0026gt;\u0026thinsp;3mm), the presence of fecaliths in the appendiceal lumen, and the presence of surrounding fluid. While, the pathological and intraoperative records withhold verifying the accuracy of the clinical diagnosis based on the postoperative pathological diagnosis and surgical exploration results. The data were extracted through a structured electronic medical record system to ensure the integrity and consistency of the information, providing comprehensive data support for subsequent multivariate analysis.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e\u003ch2\u003e2.3 Data processing, and construction of ML prediction models\u003c/h2\u003e\u003cp\u003eAfter completing the clinical data collection in the preliminary stage of the study, the integrity and accuracy of the data were first checked. Records with obvious logical errors were removed, and the missing values were filled using K-nearest neighbors (KNN) algorithm [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e] which screens the nearest neighbor samples by calculating the Euclidean distance between samples, and performs data imputation based on the similarity features.\u003c/p\u003e\u003cp\u003eFor dataset splitting, a fixed random seed strategy was adopted by dividing the total dataset into a training set (70%) and a test set (30%) at a ratio of 7:3. Lasso regression (Least Absolute Shrinkage and Selection Operator) was applied for high-dimensional feature selection on the data of the training set, which introduced the L1 regularization constraint condition to compress the coefficients of redundant features to zero, and finally selected a feature subset that was significantly correlated with the target variable. The ML prediction models comprised of logistic Regression (LR) with odd ration (OR), XGBoost, support vector machine (SVM) with selection of kernel functions (such as the Gaussian kernel and the polynomial kernel), and random forest (RF) models as reported previously [\u003cspan additionalcitationids=\"CR24 CR25 CR26\" citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eA multi-dimensional verification strategy was adopted to systematically evaluate the generalization ability and clinical applicability of the prediction model. Firstly, we constructed the Receiver Operating Characteristic (ROC) curve to quantify the model's ability to distinguish gangrene and perforation events, and calculated the area under the curve (AUC) as the core evaluation index. In addition, we also supplemented it with classification performance indicators such as accuracy, sensitivity, and specificity. The DeLong non-parametric test method [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e] was used for the difference test of the ROC curves among the models. Then, we visualized the consistency between the model's predicted probability and the actual observed probability through the calibration curve, and combined the Hosmer-Lem show test to quantify the goodness of calibration (P\u0026thinsp;\u0026gt;\u0026thinsp;0.05 indicates that the model is well calibrated). This calibration curve shows the degree of coincidence between the predicted risk and the true risk distribution in the form of quantile grouping. Finally, in the evaluation of the degree of risk and benefit, we used Decision Curve Analysis (DCA) to quantify the clinical net benefit of the model under different risk thresholds, to evaluate the practical application value of the model.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e\u003ch2\u003e2.4 Statistical analysis\u003c/h2\u003e\u003cp\u003eThis study systematically analyzed 38 clinical characteristics of 251 patients based on the R statistical platform (version number 4.4.2; access website: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.R-project.org\u003c/span\u003e\u003cspan address=\"https://www.R-project.org\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). In the data preprocessing stage, firstly, the Kolmogorov-Smirnov test was used to evaluate the normal distribution characteristics of continuous variables: for the indicators that conform to the normal distribution, the independent samples T-test was adopted, and the results were presented as mean\u0026thinsp;\u0026plusmn;\u0026thinsp;standard deviation [\u003cspan class=\"InlineEquation\"\u003e\u003c/span\u003e]. For the data with non-normal distribution, the Mann-Whitney rank sum test was used, and it was described by the median and interquartile range (first quartile, third quartile) [M(P25, P75)]. Categorical variables were analyzed by the chi-square test, and the frequency distribution was expressed as the number of cases and the proportion [N(%)]. All statistical analyses used a two-sided P value\u0026thinsp;\u0026lt;\u0026thinsp;0.05 as the criterion for judging that the difference was statistically significant. The R packages used were: 'caret', 'tidyverse', 'autoReg', 'ggplot2', 'compareGroups', 'Table\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e', 'plyr', 'corrplot', 'glmnet', 'rrtable', 'Hmisc','reportROC', 'rmda', 'randomForest', 'dplyr', 'rms', 'data.table', 'xgboost', 'ggpubr', 'Matrix', 'e1071', 'nortest', 'plotly', 'Ckmeans.1d.dp', 'ggprism', 'CBCgrps', 'DiagrammeR','shapviz', and 'pROC'.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eComparison results of general clinical characteristics on two groups\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"7\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eCharacteristics\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"3\" nameend=\"c4\" namest=\"c2\"\u003e\u003cp\u003eTraining Set (n\u0026thinsp;=\u0026thinsp;177)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"3\" nameend=\"c7\" namest=\"c5\"\u003e\u003cp\u003eTest Set (n\u0026thinsp;=\u0026thinsp;74)\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eNon-gangrene and Perforation\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eGangrene and Perforation\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eP-value\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eNon-gangrene and Perforation\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eGangrene and Perforation\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u003cp\u003eP-value\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePatients (n)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e128\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e49\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e54\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e20\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge (years) (IQR)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e68 (64,73.25)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e68 (64,75)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.654\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e68 (64,71)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e68.5 (64.75,74.25)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.241\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eWeight (kg)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e60.47\u0026thinsp;\u0026plusmn;\u0026thinsp;8.97\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e57.55\u0026thinsp;\u0026plusmn;\u0026thinsp;9.78\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.060\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e59.74\u0026thinsp;\u0026plusmn;\u0026thinsp;10.23\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e61.25\u0026thinsp;\u0026plusmn;\u0026thinsp;11.96\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.62\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHeight (cm)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e161.09\u0026thinsp;\u0026plusmn;\u0026thinsp;7.65\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e160.76\u0026thinsp;\u0026plusmn;\u0026thinsp;7.01\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.788\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e160.81\u0026thinsp;\u0026plusmn;\u0026thinsp;9.02\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e161.4\u0026thinsp;\u0026plusmn;\u0026thinsp;5.29\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.731\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eBMI (kg/m\u003csup\u003e2\u003c/sup\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e23.32\u0026thinsp;\u0026plusmn;\u0026thinsp;3.25\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e22.2\u0026thinsp;\u0026plusmn;\u0026thinsp;3.03\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.034\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e23.09\u0026thinsp;\u0026plusmn;\u0026thinsp;3.57\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e23.56\u0026thinsp;\u0026plusmn;\u0026thinsp;4.70\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.645\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colspan=\"7\" nameend=\"c7\" namest=\"c1\"\u003e\u003cp\u003e\u003cb\u003eGender (n, %)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFemale\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e72(56)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e22(45)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.236\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e30(56)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e9(45)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.585\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMale\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e56(44)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e27(55)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e24(44)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e11(55)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colspan=\"7\" nameend=\"c7\" namest=\"c1\"\u003e\u003cp\u003e\u003cb\u003eDiabetes (n, %)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e115(90)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e45(92)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.783\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e47(87)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e18(90)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eYes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e13(10)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e4(8)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e7(13)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e2(10)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colspan=\"7\" nameend=\"c7\" namest=\"c1\"\u003e\u003cp\u003e\u003cb\u003eHypertension (n, %)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e80 (62)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e35 (71)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.348\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e37 (69)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e9 (45)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.113\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eYes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e48 (38)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e14 (29)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e17 (31)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e11 (55)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colspan=\"7\" nameend=\"c7\" namest=\"c1\"\u003e\u003cp\u003e\u003cb\u003eSmoking (n, %)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e117(91)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e37(76)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.01\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e50(93)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e16(80)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.2\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eYes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e11(9)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e12(24)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e4(7)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e4(20)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colspan=\"7\" nameend=\"c7\" namest=\"c1\"\u003e\u003cp\u003e\u003cb\u003eDrinking (n, %)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e120(94)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e45(92)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.74\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e53(98)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e18(90)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.176\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eYes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e8(6)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e4(8)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1(2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e2(10)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colspan=\"7\" nameend=\"c7\" namest=\"c1\"\u003e\u003cp\u003e\u003cb\u003eTime from Onset to Surgery (d) (IQR)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1(1,2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1(1,3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1(1,2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1(1,2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.781\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colspan=\"7\" nameend=\"c7\" namest=\"c1\"\u003e\u003cp\u003e\u003cb\u003eUse of Antibacterial Drugs (n, %)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e70(55)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e23(47)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.45\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e27(50)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e10(50)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eYes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e58(45)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e26(53)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e27(50)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e10(50)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colspan=\"7\" nameend=\"c7\" namest=\"c1\"\u003e\u003cp\u003e\u003cb\u003eMigratory Right Lower Quadrant Pain (n, %)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eYes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e128(100)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e49(100)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e54(100)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e20(100)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colspan=\"7\" nameend=\"c7\" namest=\"c1\"\u003e\u003cp\u003e\u003cb\u003eNausea and Vomiting (n, %)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e64(50)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e28(57)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.495\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e31(57)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e9(45)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.491\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eYes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e64(50)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e21(43)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e23(43)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e11(55)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colspan=\"7\" nameend=\"c7\" namest=\"c1\"\u003e\u003cp\u003e\u003cb\u003eAnorexia (n, %)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e28(22)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e7(14)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.356\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e12(22)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e2(10)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.326\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eYes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e100(78)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e42(86)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e42(78)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e18(90)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colspan=\"7\" nameend=\"c7\" namest=\"c1\"\u003e\u003cp\u003e\u003cb\u003eFever (n, %)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e114(89)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e32(65)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e49(91)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e15(75)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.122\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eYes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e14(11)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e17(35)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e5(9)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e5(25)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colspan=\"7\" nameend=\"c7\" namest=\"c1\"\u003e\u003cp\u003e\u003cb\u003eTenderness in Right Lower Quadrant (n, %)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e7(5)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1(2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.447\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e4(7)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e2(10)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.659\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eYes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e121(95)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e48(98)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e50(93)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e18(90)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colspan=\"7\" nameend=\"c7\" namest=\"c1\"\u003e\u003cp\u003e\u003cb\u003eRebound Tenderness in Right Lower Quadrant (n, %)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e33(26)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e12(24)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e18(33)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e5(25)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.685\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eYes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e95(74)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e37(76)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e36(67)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e15(75)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003c/div\u003e"},{"header":"3 Results","content":"\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003e3.1 General clinical characteristics\u003c/h2\u003e\u003cp\u003eA total of n\u0026thinsp;=\u0026thinsp;251 elderly patients who underwent laparoscopic appendectomy were included as per selection criteria and the general clinical characteristics of the research subjects are presented in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. The patients (n\u0026thinsp;=\u0026thinsp;251) with acute appendicitis were divided into the gangrene and perforation group (n\u0026thinsp;=\u0026thinsp;69) and the non-gangrene and perforation group (n\u0026thinsp;=\u0026thinsp;182). Among them, the training set (n\u0026thinsp;=\u0026thinsp;177) included n\u0026thinsp;=\u0026thinsp;49 patients with gangrene and perforation and n\u0026thinsp;=\u0026thinsp;128 patients without gangrene and perforation; the test set (n\u0026thinsp;=\u0026thinsp;74) included n\u0026thinsp;=\u0026thinsp;20 patients with gangrene and perforation and n\u0026thinsp;=\u0026thinsp;54 patients without gangrene and perforation.\u003c/p\u003e\u003cp\u003eThe results showed that in the training sample set of this study, the gender distribution in the gangrene and perforation group for male and female was 55% (n\u0026thinsp;=\u0026thinsp;27) and 45% (n\u0026thinsp;=\u0026thinsp;22), respectively. The average age of the patients was 68 years old. In terms of comorbidities, history of diabetes, hypertension, smoking, and drinking was ,8% (n\u0026thinsp;=\u0026thinsp;4), 29% (n\u0026thinsp;=\u0026thinsp;14), 24% (n\u0026thinsp;=\u0026thinsp;12), and 8% (n\u0026thinsp;=\u0026thinsp;4), respectively. The average time from the onset of the disease to the operation was 1 day, while 53% (n\u0026thinsp;=\u0026thinsp;26) patients had a history of taking antibacterial drugs before the onset of the disease. All patients showed migratory right lower quadrant pain. A total of 43% (n\u0026thinsp;=\u0026thinsp;21) patients had symptoms of nausea and vomiting, 86% (n\u0026thinsp;=\u0026thinsp;42) patients had anorexia, 35%(n\u0026thinsp;=\u0026thinsp;17) patients had a history of fever, 98%(n\u0026thinsp;=\u0026thinsp;48) patients had tenderness in the right lower quadrant, and 76% (n\u0026thinsp;=\u0026thinsp;37) patients had rebound tenderness in the right lower quadrant (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e)While, in the non-gangrene and perforation group, there were 44%(n\u0026thinsp;=\u0026thinsp;56) male and 56% (n\u0026thinsp;=\u0026thinsp;72)female patients. The average age of the patients was 68 years old. Patients with history of diabetes, hypertension, smoking, and drinking were 10% (n\u0026thinsp;=\u0026thinsp;13), 38% (n\u0026thinsp;=\u0026thinsp;48), 9% (n\u0026thinsp;=\u0026thinsp;11) and 7% (12), respectively. The average time from the onset of the disease to the operation was 1 day; while 45% (n\u0026thinsp;=\u0026thinsp;58) patients had a history of taking antibacterial drugs. All patients had migratory right lower quadrant pain. A total of 50% (n\u0026thinsp;=\u0026thinsp;64) patients with nausea and vomiting78%(n\u0026thinsp;=\u0026thinsp;100) patients with anorexia, 11% (n\u0026thinsp;=\u0026thinsp;14) patients with a history of fever, 95% (n\u0026thinsp;=\u0026thinsp;121) patients with tenderness in the right lower quadrant, and 74% (n\u0026thinsp;=\u0026thinsp;95) patients with rebound tenderness in the right lower quadrant were identified (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eIn the training set, there were significant statistical differences in BMI (P\u0026thinsp;=\u0026thinsp;0.034), smoking history (P\u0026thinsp;=\u0026thinsp;0.01), and fever history (P\u0026thinsp;\u0026lt;\u0026thinsp;0.001) between the gangrene and perforation group and the non-gangrene and perforation group; while in the test set, no statistically significant differences in characteristics were found between the gangrene and perforation group and the non-gangrene and perforation group.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e\u003ch2\u003e3.2 Preoperative laboratory tests\u003c/h2\u003e\u003cp\u003eThe results of preoperative laboratory tests (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e) showed that In training set there were significant differences between the gangrene and perforation group and non-gangrene and perforation group in terms of WBCs (P\u0026thinsp;=\u0026thinsp;0.001), neutrophils (P\u0026thinsp;=\u0026thinsp;0.004), lymphocytes (P\u0026thinsp;\u0026lt;\u0026thinsp;0.001), CRP (P\u0026thinsp;\u0026lt;\u0026thinsp;0.001), procalcitonin (P\u0026thinsp;\u0026lt;\u0026thinsp;0.001), D-dimer (P\u0026thinsp;\u0026lt;\u0026thinsp;0.001), indirect bilirubin (P\u0026thinsp;=\u0026thinsp;0.03), Alb (P\u0026thinsp;\u0026lt;\u0026thinsp;0.001), and serum Na⁺ level (P\u0026thinsp;=\u0026thinsp;0.012). While, in the test set, the WBC (P\u0026thinsp;=\u0026thinsp;0.016), neutrophils (P\u0026thinsp;=\u0026thinsp;0.027), lymphocytes (P\u0026thinsp;=\u0026thinsp;0.014), hemoglobin (P\u0026thinsp;=\u0026thinsp;0.036), hematocrit (P\u0026thinsp;=\u0026thinsp;0.02), CRP (P\u0026thinsp;\u0026lt;\u0026thinsp;0.001), procalcitonin (P\u0026thinsp;=\u0026thinsp;0.009), D - dimer (P\u0026thinsp;=\u0026thinsp;0.032), and Alb (P\u0026thinsp;=\u0026thinsp;0.037).\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eComparison results of preoperative laboratory tests on two groups\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"7\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eCharacteristics\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"3\" nameend=\"c4\" namest=\"c2\"\u003e\u003cp\u003eTraining Set (n\u0026thinsp;=\u0026thinsp;177)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"3\" nameend=\"c7\" namest=\"c5\"\u003e\u003cp\u003eTest Set (n\u0026thinsp;=\u0026thinsp;74)\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eNon-gangrene and Perforation\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eGangrene and Perforation\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eP\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eNon-gangrene and Perforation\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eGangrene and Perforation\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u003cp\u003eP\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePatients (n)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e128\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e49\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e54\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e20\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePLT (*10^9/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e217.16\u0026thinsp;\u0026plusmn;\u0026thinsp;65.51\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e203.33\u0026thinsp;\u0026plusmn;\u0026thinsp;54.16\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.155\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e200.31\u0026thinsp;\u0026plusmn;\u0026thinsp;54.02\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e201.65\u0026thinsp;\u0026plusmn;\u0026thinsp;72.74\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.941\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eWBC (*10^9/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e11.18\u0026thinsp;\u0026plusmn;\u0026thinsp;3.85\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e13.64\u0026thinsp;\u0026plusmn;\u0026thinsp;4.56\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e11.98\u0026thinsp;\u0026plusmn;\u0026thinsp;3.18\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e14.36\u0026thinsp;\u0026plusmn;\u0026thinsp;4.82\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.016\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNEUT (%)(IQR)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e83.65\u003c/p\u003e\u003cp\u003e(73.65,88.73)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e87(82.6,91)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.004\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e83.85(78.1,87.2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e86.55\u003c/p\u003e\u003cp\u003e(83.03,91.7)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.027\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLYM (%)(IQR)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e11.5(7.25,19.7)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e8.3(4.9,11.5)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e11.9(7.53,15.55)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e7.9(4.47,11.03)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.014\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHGB (g/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e144.62\u0026thinsp;\u0026plusmn;\u0026thinsp;14.69\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e146.92\u0026thinsp;\u0026plusmn;\u0026thinsp;19.68\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.46\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e146.72\u0026thinsp;\u0026plusmn;\u0026thinsp;14.82\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e137.75\u0026thinsp;\u0026plusmn;\u0026thinsp;18.95\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.036\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHCT (%)(IQR)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.43(0.41,0.46)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.43(0.4,0.48)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.51\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.44(0.42,0.46)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.41(0.39,0.44)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.02\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMCV (fL)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e92.08\u0026thinsp;\u0026plusmn;\u0026thinsp;4.34\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e92.66\u0026thinsp;\u0026plusmn;\u0026thinsp;4.59\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.448\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e92.19\u0026thinsp;\u0026plusmn;\u0026thinsp;4.63\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e90.48\u0026thinsp;\u0026plusmn;\u0026thinsp;6.33\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.205\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMCHC (g/L)(IQR)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e337\u003c/p\u003e\u003cp\u003e(331.75,342)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e339(328,347)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.585\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e334.5\u003c/p\u003e\u003cp\u003e(331.25,340)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e334.5\u003c/p\u003e\u003cp\u003e(330.75,343)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.63\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRDW (fL)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e44.03\u0026thinsp;\u0026plusmn;\u0026thinsp;3.02\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e44.27\u0026thinsp;\u0026plusmn;\u0026thinsp;3.42\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.648\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e43.62\u0026thinsp;\u0026plusmn;\u0026thinsp;2.53\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e44.24\u0026thinsp;\u0026plusmn;\u0026thinsp;3.04\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.417\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMPV (fL)(IQR)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e9.9(9.17,10.7)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e10(9.4,10.4)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.887\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e10.4(9.8,11.38)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e10(9.35,11)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.281\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePCT (\u0026micro;g/L)(IQR)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.11(0.06,0.32)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.84(0.19,7.62)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.09(0.06,0.7)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1.41(0.08,40.73)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.009\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCRP (mg/L)(IQR)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e42.34\u003c/p\u003e\u003cp\u003e(13.59,70.15)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e91.39\u003c/p\u003e\u003cp\u003e(44.21,142.11)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e33.14\u003c/p\u003e\u003cp\u003e(10.8,63.51)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e98.88\u003c/p\u003e\u003cp\u003e(36.4,136.29)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eD-Dimer (\u0026micro;g/L)(IQR)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.51(0.34,0.99)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1.08(0.66,1.98)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.54(0.39,1.13)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1.61(0.55,5.11)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.032\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDBIL(\u0026micro;mol/L)(IQR)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e4.9(3.48,7)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e5.5(3.6,8.1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.231\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e4.3(2.92,6.05)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e5.7(3.33,8.75)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.154\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eIBIL(\u0026micro;mol/L)(IQR)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e15.25\u003c/p\u003e\u003cp\u003e(11.1,22.45)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e19.9(12.9,27.9)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.03\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e16.95(10.38,23)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e15.15\u003c/p\u003e\u003cp\u003e(11.78,19.62)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.831\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eALB(g/L)(IQR)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e40.65\u003c/p\u003e\u003cp\u003e(38.48,42.73)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e38.6(35.7,41.2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e40.55\u003c/p\u003e\u003cp\u003e(38.6,42.42)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e38.8\u003c/p\u003e\u003cp\u003e(35.77,40.62)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.037\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNa\u003csup\u003e+\u003c/sup\u003e(mmol/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e136.09\u0026thinsp;\u0026plusmn;\u0026thinsp;3.10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e132.65\u0026thinsp;\u0026plusmn;\u0026thinsp;14.53\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.012\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e136.35\u0026thinsp;\u0026plusmn;\u0026thinsp;2.43\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e134.86\u0026thinsp;\u0026plusmn;\u0026thinsp;3.65\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.104\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colspan=\"7\" nameend=\"c7\" namest=\"c1\"\u003e\u003cp\u003eAbbreviations: CRP\u0026thinsp;=\u0026thinsp;C-reactive proteins; WBCs\u0026thinsp;=\u0026thinsp;white blood cells\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\u003ch2\u003e3.3 Preoperative imaging examinations\u003c/h2\u003e\u003cp\u003eIn terms of imaging examinations (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e) there were statistically significant differences in all characteristics between the non-gangrene and perforation group and the gangrene and perforation group in the training set. While, only the minimum diameter of the appendix (P\u0026thinsp;=\u0026thinsp;0.001) in the test set, showed significant difference between the non-gangrene and perforation group and the gangrene and perforation group.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eComparison results of preoperative imaging examinations on two groups\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"7\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eCharacteristics\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"3\" nameend=\"c4\" namest=\"c2\"\u003e\u003cp\u003eTraining Set (n\u0026thinsp;=\u0026thinsp;177)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"3\" nameend=\"c7\" namest=\"c5\"\u003e\u003cp\u003eTest Set (n\u0026thinsp;=\u0026thinsp;74)\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eNon-gangrene and Perforation\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eGangrene and Perforation\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eP-value\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eNon-gangrene and Perforation\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eGangrene and Perforation\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u003cp\u003eP-value\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePatients (n)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e128\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e49\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e54\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e20\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMinimum diameter of the appendix (cm) (IQR)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.9(0.7,1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.9(0.8,1.1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.04\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.8(0.76,1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1.2(0.98,1.25)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.001\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eAppendiceal wall thickening (n, %)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e0.004\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e\u003cb\u003e0.569\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e18(14)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0(0)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e4(7)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0(0)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eYes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e110(86)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e49(100)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e50(93)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e20(100)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eFecalith in the appendiceal lumen (n, %)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e0.049\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e\u003cb\u003e1\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e66(52)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e34(69)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e31(57)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e11(55)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eYes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e62(48)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e15(31)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e23(43)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e9(45)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eFluid around the appendix (n, %)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003e0.025\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e\u003cb\u003e0.207\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNo\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e28(22)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3(6)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e14(26)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e2(10)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eYes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e100(78)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e46(94)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e40(74)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e18(90)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003e3.4 Prediction potential of four machine learning models\u003c/h2\u003e\u003cp\u003eWe next evaluated the prediction potential of four ML models and results indicated that during the training stage the AUC value of RF reached 0.999 (95% CI, 0.998\u0026ndash;1.000), which was significantly higher than XGBoost (0.975), LR (0.774), and SVM (0.768) models (Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). Nevertheless, the values for all models were higher than 0.7, which pretends that all four prediction models have good prediction potential in the training environment. Contrarily, the models once transferred to the test set showed obvious differentiation in their performances: LR became the best-performing model with an AUC of 0.768 (95% CI, 0.642\u0026ndash;0.893), followed by SVM (0.751), XGBoost (0.725), and RF (0.686) in descending order. Among them, the performance of RF decreased significantly. Further analysis of other indicators showed that LR had an advantage in accuracy (0.784) and specificity (0.741), while XGBoost had the highest sensitivity of 0.836, but its specificity was only 0.600. Although the accuracies of SVM and RF were 0.757, but their sensitivities were above 0.7.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eFour ML prediction models\u0026rsquo; outcomes\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"6\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eModel\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAUC\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eAccuracy\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eSensitivity\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eSpecificity\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003e95%\u003cem\u003eCI\u003c/em\u003e\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLR\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.768\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.784\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.700\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.741\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.642,0.893\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eXGBoost\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.725\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.773\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.836\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.600\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.584,0.867\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSVM\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.751\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.757\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.773\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.625\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.620,0.882\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRF\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.686\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.757\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.810\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.563\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.533,0.839\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eIn this study, when 177 cases in the training set were included for statistical analysis, whether the patient was diagnosed with appendiceal gangrene and perforation was used as the outcome indicator for model building. The Lasso regression algorithm was used for feature dimensionality reduction to eliminate the influence of multicollinearity among variables, and finally, the characteristic variables with significant predictive value were screened out. Figure\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e objectively demonstrates the dynamic adjustment process of the model parameters' shrinkage under different λ values.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eIn this study, regression models of feature subsets, based on λmin and λ1se were constructed simultaneously. To moderately relax the restrictions on the model complexity and avoid overfitting the noisy data, this study selected to include the 9 features corresponding to the λ1se value within one standard error of the model error for the subsequent construction of the machine learning prediction model.\u003c/p\u003e\u003cp\u003eThe features screened by the Lasso regression include 9 features: smoking history, fever history, WBC, percentage of lymphocytes, CRP, PCT, Alb, serum Na\u0026thinsp;+\u0026thinsp;level, and whether the appendiceal wall is thickened.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003e3.5 Model interpretation and individual analysis\u003c/h2\u003e\u003cp\u003eFirstly, we conducted univariate and multivariate LR analyses on the 9 features screened out by the Lasso regression respectively (Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). The results indicated that all the 9 features showed significant differences between the gangrene and perforation group and the non-gangrene and perforation group in the univariate logistic regression analysis, and among them, 3 features were identified as independent risk factors in the multivariate logistic regression analysis (Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). The three independent risk factors are: WBC (OR\u0026thinsp;=\u0026thinsp;1.12, 95% CI, 1.01\u0026ndash;1.25, P\u0026thinsp;=\u0026thinsp;0.038), CRP (OR\u0026thinsp;=\u0026thinsp;1.01, 95% CI, 1.00\u0026ndash;1.01, P\u0026thinsp;=\u0026thinsp;0.041), and Alb (OR\u0026thinsp;=\u0026thinsp;0.85, 95% CI, 0.76\u0026ndash;0.95, P\u0026thinsp;=\u0026thinsp;0.004).\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab5\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eSingle and multiple factors LR results of nine characteristics\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"5\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eCharacteristics\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e\u003cp\u003eSingle factors LR\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003eMultiple factors LR\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eOR (95%\u003cem\u003eCI)\u003c/em\u003e\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eP-value\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eOR (95%\u003cem\u003eCI)\u003c/em\u003e\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eP-vlaue\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSmoking\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e3.45 (1.41\u0026ndash;8.47)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.007\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e2.72 (0.90\u0026ndash;8.20)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.076\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFever\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e4.33 (1.93\u0026ndash;9.71)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e2.03 (0.74\u0026ndash;5.51)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.167\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eWBC\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.16 (1.06\u0026ndash;1.26)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1.12 (1.01\u0026ndash;1.25)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.038\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLYM%\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.91 (0.86\u0026ndash;0.97)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.002\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.95 (0.88\u0026ndash;1.03)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.221\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePCT\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.04 (1.01\u0026ndash;1.08)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.007\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1.01 (0.97\u0026ndash;1.05)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.640\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCRP\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.01 (1.01\u0026ndash;1.02)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1.01 (1.00-1.01)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.041\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eALB\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.84 (0.76\u0026ndash;0.92)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.85 (0.76\u0026ndash;0.95)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.004\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNa+\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.87 (0.78\u0026ndash;0.97)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.011\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.94 (0.85\u0026ndash;1.03)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.194\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAppendiceal wall thickening\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.986\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eAccording to the regression coefficients of the features (Table\u0026nbsp;\u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003e6\u003c/span\u003e), our LR model is calculated using the following formula: logit(Y) = -2.0562\u0026thinsp;+\u0026thinsp;0.1086*WBC\u0026thinsp;+\u0026thinsp;0.0142*CRP \u0026minus;\u0026thinsp;0.0335*Alb (binary predictive features take values of 0 or 1 in the formula).\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab6\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 6\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eThree characteristics of the LR model\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"5\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCharacteristics\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eCoefficient\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eStandard error\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eZ-statistics\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003ePr(\u0026gt;|Z|)\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eIntercept\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-2.0562\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3.1997\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-0.64\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.5205\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eWBC\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.1086\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.0829\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1.31\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.1901\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePCT\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.0142\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.0060\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e2.38\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.0171\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eALB\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-0.0335\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.0752\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-0.45\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.6560\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eThe XGBoost prediction framework uses the gradient - boosted decision tree ensemble architecture to achieve modeling optimization. Its core mechanism is to build weak learners iteratively in stages. During each tree-structure splitting process, the information gain index for splitting nodes was calculated based on second-order Taylor expansion of the loss function, which is used to quantify the marginal improvement effect of candidate features on the model performance. By recursively accumulating the splitting gain values of each feature in all decision trees, the system generates global weighted statistics of multi-dimensional predictive variables and finally constructs a hierarchical feature importance evaluation system.\u003c/p\u003e\u003cp\u003eTo interpret the effect of specified features, we applied SHAP and the summary plot showed that CRP has the highest prediction importance among clinical features. Its SHAP value is significantly higher than that of other features, indicating that CRP has the greatest global impact on the model output, while the other laboratory indicators such as Alb and WBC follow closely.\u003c/p\u003e\u003cp\u003eDuring the training stage of the XGBoost prediction model, its AUC reached 0.975 (95% CI, 0.953\u0026ndash;0.998), with an accuracy rate of 93.8%. Meanwhile, the sensitivity and specificity were maintained at 93.9%, and 93.2%, respectively. While, in the test set, the AUC was 0.725 (95% CI, 0.584\u0026ndash;0.867), and the overall prediction accuracy rate dropped to 77.3%. Meanwhile, the sensitivity index remained at 83.6%, but the specificity decreased significantly to 60.0%.\u003c/p\u003e\u003cp\u003eThe method of SVM algorithm involves introducing features with different ranks one by one with CRP, WBC, and Alb in sequence. The choice of kernel function has a decisive impact on the performance of the SVM model. By comparing the AUCs of four linear kernel functions-Linear, Polynomial, Radial, and Sigmoid-on the test set, were 0.751, 0.671, 0.663, and 0.588 respectively. Based on the linear kernel function with the highest AUC value (Linear Kernel), we constructed the SVM model and obtained a three-dimensional decision boundary graph of the training set (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eDuring the training stage the AUC of the SVM prediction model reached 0.768 (95%Cl, 0.684\u0026ndash;0.852), with 76.8% accuracy and 77% sensitivity with 75% specificity. In the test set, the model maintained an AUC of 0.751 (95%Cl, 0.620\u0026ndash;0.882), with slightly decreased accuracy rate (75.7%). Meanwhile, the sensitivity increased to 77.3%, but the specificity decreased to 62.5%.\u003c/p\u003e\u003cp\u003eNext, we evaluated the contribution intensity of features to optimize the decision path during the node splitting process (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). The MDA analysis showed the ranking as WBC (13.306), CRP (8.880), and Alb (4.430). Similar trend was observed in the MDG analysis as WBC (24.255), CRP (23.788), and Alb (21.868) in sequence.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eWe further selected 13 decision trees with the lowest classification error rate to construct the RF model. During the training stage, the model shows an AUC of up to 0.999 (95%Cl, 0.998\u0026ndash;1.000), with an accuracy rate of 99.4%, and maintained a sensitivity and specificity of 99.2% and 100%, respectively. While, in the test set, the model's performance undergoes a significant drop in AUC (0.686 (95%Cl, 0.533\u0026ndash;0.839)), accuracy (75.7%), sensitivity (81.0%), and the specificity (56.3%).\u003c/p\u003e\u003cp\u003eThe ROC curve based on the training set shows (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e) that there are significant differences in the AUCs of the four machine learning methods. Specifically, the RF model demonstrates excellent performance in the classification task (AUC 0.999, 95% CI, 0.998\u0026ndash;1.000), and its prediction performance is superior to that of XGBoost (AUC 0.975, 95% CI, 0.953\u0026ndash;0.998), LR (AUC 0.774, 95% CI, 0.692\u0026ndash;0.856), and SVM (AUC 0.768, 95% CI, 0.684\u0026ndash;0.852). The AUCs of all models exceeded 0.700, indicating that the prediction performances of these models on the training set are good.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eIn the test dataset, the ROC curves of the four prediction models are shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e. Among them, the LR model performs the best with an AUC value of 0.768 (95% CI, 0.642\u0026ndash;0.893) as compared to SVM is 0.751 (95% CI, 0.620\u0026ndash;0.882), XGBoost is 0.725 (95% CI, 0.584\u0026ndash;0.867), and RF is 0.686 (95% CI, 0.533\u0026ndash;0.839), pretending RF model exhibits more significant characteristics of performance attenuation.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\u003ch2\u003e3.6 Calibration analysis of LR model\u003c/h2\u003e\u003cp\u003eIn this study, the LR model was finally determined as the optimal prediction tool. The verification results showed that the model demonstrated excellent comprehensive performance on the test set, with an AUC of 0.768 (95% CI, 0.642\u0026ndash;0.893), sensitivity and specificity of 0.700, and 0.741, respectively of which were significantly better than other models. To further evaluate the reliability of the model, we used the Bootstrap resampling method (1000 repetitions) for internal calibration verification (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e). The results showed that the calibration curves of the training set and the test set were highly consistent with the ideal reference line (y\u0026thinsp;=\u0026thinsp;x), with an average absolute error of 0.023, and there was no statistical difference in the Hosmer-Lem show test (P\u0026thinsp;=\u0026thinsp;0.67), confirming that there was a good consistency between the model's predicted probability and the actual occurrence probability.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\u003ch2\u003e3.7 Clinical applicability analysis of LR model\u003c/h2\u003e\u003cp\u003eNext, we systematically evaluated the clinical application value of the LR model in elderly patients with gangrenous perforation of acute appendicitis by plotting the DCA curve of the LR model (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003e). Results showed that the DCA curve of the LR model is always higher than the reference lines of the extreme strategies of \"full intervention\" and \"no intervention\". This characteristic of the curve distribution indicates that this model is used to guide clinical decision-making by generating a significant net clinical benefits within a wide range of risk probability intervals.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e\u003ch2\u003e3.8 Nomogram of LR model\u003c/h2\u003e\u003cp\u003eThe LR model was then transformed into a visual clinical tool by constructing a Nomogram (Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003e), which integrated the combined predictive effects of three biomarkers, namely WBC, CRP, and Alb. Based on the quantitative framework of the synergistic effect of multiple indicators, this tool intuitively displays the risk contribution weights of each indicator through scale line segments in a two-dimensional coordinate system. Gastrointestinal surgeons can quickly read the scores of each indicator of the patient through the horizontal axis, obtain the total score after vertical accumulation, and finally locate the corresponding probability of gangrenous perforation on the risk probability axis.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e"},{"header":"4. Discussion","content":"\u003cp\u003eAcute appendicitis in the elderly tends to progress more rapidly and severely than younger individuals and this accelerated progression is attributed to the unique pathophysiology[\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. When the lumen of the appendix is obstructed, bacteria proliferate within the closed environment, triggering a severe inflammatory response which in turn impairs the blood supply to the appendix, leading to gangrene and perforation, hereby exacerbating the inflammatory cascade[\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]. Clinically, this is manifested as initial periumbilical pain that migrates to the right lower quadrant, accompanied by gastrointestinal symptoms such as nausea, vomiting, and abdominal distension, as well as systemic symptoms like fever[\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]. However, the disease progresses more rapidly in elderly patients due to diminished physiological functions and weakened inflammatory response. Once gangrene and perforation occur, the spread of inflammation can easily lead to severe peritonitis, extensive intra-abdominal infection, and even septic shock, severely compromising the function of multiple organ systems[\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]. Given the vulnerability of elderly patients to prolonged illness, it is crucial to ensure the timely medical intervention.\u003c/p\u003e\u003cp\u003eThe diagnosis of gangrenous perforation in elderly acute appendicitis is clinically challenging due to frequently atypical symptoms, such as mild abdominal pain, minimal gastrointestinal reactions, and less pronounced fever, which can easily be ignored by both patients and clinicians[\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e]. Laboratory tests including WBC count and neutrophil percentage may not show significant elevation, often reflecting the blunted inflammatory response is in this population. The malposition of appendix or intra-abdominal gas often interfere the imaging modalities such as abdominal ultrasound hereby limiting the diagnosis of appendiceal gangrene and perforation. Although CT is highly valuable for diagnosis, however, the health conditions of some elderly patients may not tolerate the procedure. Furthermore, the presence of multiple comorbidities in elderly patients can mask the symptoms of appendiceal gangrene and perforation thereby increasing the likelihood of misdiagnosis or delayed diagnosis[\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eML has emerged as a powerful tool in the development of diagnostic models for various diseases, demonstrating the superior performance compared to traditional statistical methods[\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]. Unlike conventional approaches, ML enables computers to \"learn\" from data without explicit programming, allowing the efficient handling of large datasets through complex interactions. By uncovering intricate patterns and associations within the data, ML provides the researchers with novel analytical perspectives and is increasingly playing a vital role in the medical research[\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eOur study identified WBC, CRP and Alb as independent risk factors for gangrenous perforation in elderly acute appendicitis. Utilizing ML techniques, we constructed four models\u0026mdash;LR, RF, XGBoost, and SVM\u0026mdash;and evaluated their performance using metrics such as AUC, sensitivity, and specificity. In the training dataset, the RF model exhibited excellent discriminative ability with an AUC of 0.999, significantly outperforming the XGBoost model (AUC\u0026thinsp;=\u0026thinsp;0.975) and the LR model (AUC\u0026thinsp;=\u0026thinsp;0.774). The superior performance of the RF model may be attributed to its ensemble of multiple decision trees, which allows it to capture complex interactions among features. However, this high degree of fit also suggests a potential risk of overfitting. In the test dataset, the LR model achieved the highest AUC (0.768), followed by the SVM model (0.751) and the XGBoost model (0.725), while the RF model's AUC dropped significantly to 0.686, indicating weaker generalizability. In terms of sensitivity and specificity, the XGBoost model had the highest sensitivity (0.836) but lower specificity (0.600), which may lead to a higher false-positive rate. In contrast, the LR model achieved a better balance between sensitivity (0.700) and specificity (0.741), making it more suitable for clinical decision-making that requires balanced control of false negatives and false positives.\u003c/p\u003e\u003cp\u003eThe LR model's DCA demonstrated net benefit within a threshold probability range of 10 to 50%, suggesting its potential utility in clinical decision-making. The LR model was further translated into a clinical tool through the development of a nomogram, which integrates the combined predictive effects of WBC, CRP, and Alb. This visual tool allows clinicians to quickly assess the risk of gangrenous perforation by summing the scores of individual biomarkers and locating the corresponding probability on the risk axis. The nomogram provides a quantitative framework for risk assessment based on the synergistic effects of multiple indicators, facilitating rapid and accurate decision-making.\u003c/p\u003e\u003cp\u003eThe LR model developed in this study is particularly suitable for rapid decision-making in regions with limited medical resources. The three biomarkers (WBC, CRP, and Alb) are widely used in primary care settings, eliminating the need for advanced imaging techniques such as CT or MRI[\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e]. This approach also reduces diagnostic costs significantly. However, the study's limitations include its single-center retrospective design and relatively small sample size, which may affect the model's generalizability. Future work should focus on validating the model through multicenter, prospective cohort studies. Additionally, efforts should be directed towards establishing a standardized database across different regions, incorporating elderly patients with diverse demographic and comorbid profiles, to enhance the model's adaptability to population heterogeneity.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThis study demonstrates that the developed machine learning model markedly enhances the diagnostic accuracy for gangrenous perforation in acute appendicitis among elderly patients, outperforming conventional analytical approaches. From an initial set of 38 clinical indicators, three key predictors\u0026mdash;WBC, CRP, and Alb\u0026mdash;were identified as critical for model development. In the RF model, WBC emerged as the most influential feature, whereas CRP was the dominant predictor in the other three models evaluated. Among the four models assessed, the LR model achieved the optimal balance of AUC, sensitivity, and specificity, despite the XGBoost model exhibiting the highest sensitivity. Given its interpretability and practical applicability in clinical settings, the LR model is considered the most effective. Furthermore, the LR model's nomogram offers an intuitive, visually accessible tool, enabling gastrointestinal surgeons to make swift and informed diagnostic decisions for elderly patients with gangrenous perforation of acute appendicitis.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003cp\u003e This study was approved by the Second Affiliated Hospital of Kunming Medical University (Approval No. PJ-2024-185). All methods were conducted in accordance with relevant guidelines and regulations, including the Declaration of Helsinki. Informed consent was not required due to the retrospective design of the study and the exemption granted by the IRB.\u003c/p\u003e\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003cp\u003eNot applicable.\u003c/p\u003e\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\u003c/p\u003e\u003ch2\u003eFunding\u003c/h2\u003e\u003cp\u003eThis study was supported by the Science and Technology Plan Project of the Department of Science and Technology of Yunnan Province (202501AY070001-030) and Yunnan Revitalization Talent Support Program (XDYC-MY-2022-0042 \u0026amp; YNWR-MY-2020-050).\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eS.F conceived the study. C.J, S.Y.B and Y.B designed the study. Y.B, L.Y.L and S.Y.B discovered the clinical effect. C.J, S.Y.Q.L and F.D analyzed the data. C.J and G.Y wrote the initial draft. S.F revised the manuscript. All authors read and approved the final version of manuscript for publication.\u003c/p\u003e\u003ch2\u003eAcknowledgements\u003c/h2\u003e\u003cp\u003eNot applicable.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eThe datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003ePowell SK. Caring for Caregivers: National Case Management Week: October 13\u0026ndash;10, 2024. Volume 29. LWW; 2024. pp. 187\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSisik A, Kudas I, Basak F, Hasbahceci M. Is the increased incidence of pathologically proven acute appendicitis more likely seen in elderly patients? A retrospective cohort study. Aging Male. 2021;24(1):1\u0026ndash;7.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKhan HT. Population ageing in a globalized world: Risks and dilemmas? J Eval Clin Pract. 2019;25(5):754\u0026ndash;60.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDhillon NK, Barmparas G, Lin T-L, Alban RF, Melo N, Yang AR, Margulies DR, Ley EJ. Unexpected complicated appendicitis in the elderly diagnosed with acute appendicitis. Am J Surg. 2019;218(6):1219\u0026ndash;22.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSegev L, Keidar A, Schrier I, Rayman S, Wasserberg N, Sadot E. Acute appendicitis in the elderly in the twenty-first century. J Gastrointest Surg. 2015;19(4):730\u0026ndash;5.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYuan J, Chen Q, Hong W, Yu L, Li X. Comparison of clinical features and outcomes of appendectomy in elderly vs. non-elderly: A systematic review and meta-analysis. Front Surg. 2022;9:818347.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eArlt A, Bharti R, Ilves I, H\u0026auml;sler R, Miettinen P, Paajanen H, Brunke G, Ellrichmann M, Rehman A, Hauser C. Characteristic changes in microbial community composition and expression of innate immune genes in acute appendicitis. Innate Immun. 2015;21(1):30\u0026ndash;41.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJung SK, Rhee DY, Lee WJ, Woo SH, Seol SH, Kim DH, Choi SP. Neutrophil-to-lymphocyte count ratio is associated with perforated appendicitis in elderly patients of emergency department. Aging Clin Exp Res. 2017;29:529\u0026ndash;36.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eOmari AH, Khammash MR, Qasaimeh GR, Shammari AK, Yaseen MKB, Hammori SK. Acute appendicitis in the elderly: risk factors for perforation. World J Emerg Surg. 2014;9:1\u0026ndash;6.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLapsa S, Ozolins A, Strumfa I, Gardovskis J. Acute appendicitis in the elderly: a literature review on an increasingly frequent surgical problem. Geriatrics. 2021;6(3):93.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFrenzel F, Hollaender S, Fries P, Stroeder R, Stroeder J. Jejunal obstruction due to rare internal hernia between skeletonized external iliac artery and vein as late complication of laparoscopic hysterectomy with pelvic lymphadenectomy\u0026mdash;case report and review of literature. Arch Gynecol Obstet. 2020;302:1075\u0026ndash;80.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCimino MM, Biloslavo A, Kurihara H, Bellio G, Porta M, Fattori S, Bass GA. Practice patterns and clinical outcomes in acute appendicitis differ in the elderly patient. Eur J Trauma Emerg Surg 2024:1\u0026ndash;10.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAl-Gaithy ZK. Clinical value of total white blood cells and neutrophil counts in patients with suspected appendicitis: retrospective study. World J Emerg Surg. 2012;7:1\u0026ndash;7.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKim JJ, Dobson BH, Ng LH, Thong C, Arthur DW, Parker T, Collaboration D, Anwari Q, Archer T, Auld L. Can normal inflammatory markers rule out acute appendicitis? The reliability of biochemical investigations in diagnosis. ANZ J Surg. 2020;90(10):1970\u0026ndash;4.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHamid MA, Afroz R, Ahmed UN, Bawani A, Khan D, Shahab R, Salim A. The importance of visualization of appendix on abdominal ultrasound for the diagnosis of appendicitis in children: A quality assessment review. World J Emerg Med. 2020;11(3):140.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePerrotta G, Geropoulos G, Bhan C. The role of imaging in the diagnosis of acute appendicitis during the COVID-19 pandemic: a retrospective cohort study. Updates Surg. 2023;75(1):205\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTulin-Silver S, Babb J, Pinkney L, Strubel N, Lala S, Milla SS, Tomita S, Fefferman NR. The challenging ultrasound diagnosis of perforated appendicitis in children: constellations of sonographic findings improve specificity. Pediatr Radiol. 2015;45:820\u0026ndash;30.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePark JH, Salminen P, Tannaphai P, Lee KH. Low-dose abdominal CT for evaluating suspected appendicitis in adolescents and young adults: review of evidence. Korean J Radiol. 2022;23(5):517.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKim D, Woodham BL, Chen K, Kuganathan V, Edye MB. Rapid MRI abdomen for assessment of clinically suspected acute appendicitis in the general adult population: a systematic review. J Gastrointest Surg. 2023;27(7):1473\u0026ndash;85.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRadak M, Lafta HY, Fallahi H. Machine learning and deep learning techniques for breast cancer diagnosis and classification: a comprehensive review of medical imaging studies. J Cancer Res Clin Oncol. 2023;149(12):10473\u0026ndash;91.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGoecks J, Jalili V, Heiser LM, Gray JW. How machine learning will transform biomedicine. Cell. 2020;181(1):92\u0026ndash;101.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSalvador\u0026ndash;Meneses J, Ruiz\u0026ndash;Chavez Z, Garcia\u0026ndash;Rodriguez J. Compressed k NN: K-nearest neighbors with data compression. Entropy. 2019;21(3):234.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBreiman L. Random forests. Mach Learn. 2001;45:5\u0026ndash;32.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHou N, Li M, He L, Xie B, Wang L, Zhang R, Yu Y, Sun X, Pan Z, Wang K. Predicting 30-days mortality for MIMIC-III patients with sepsis-3: a machine learning approach using XGboost. J translational Med. 2020;18:1\u0026ndash;14.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eStoltzfus JC. Logistic regression: a brief primer. Acad Emerg Med. 2011;18(10):1099\u0026ndash;104.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhang X, Song Q. A multi-label learning based kernel automatic recommendation method for support vector machine. PLoS ONE. 2015;10(4):e0120455.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJiang H, Ching W-K. Correlation kernels for support vector machines classification with applications in cancer data. Comput Math Methods Med. 2012;2012(1):205025.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSeshan VE, G\u0026ouml;nen M, Begg CB. Comparing ROC curves derived from regression models. Stat Med. 2013;32(9):1483\u0026ndash;93.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCohen-Arazi O, Dabour K, Bala M, Haran A, Almogy G. Management, treatment and outcomes of acute appendicitis in an elderly population: a single-center experience. Eur J Trauma Emerg Surg. 2017;43(5):723\u0026ndash;7.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTran N, Emuakhagbon VS, Baker BT, Huerta S. Re-assessing the role of the fecalith in acute appendicitis in adults: case report, case series and literature review. J Surg Case Rep. 2021;2021(1):rjaa543.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSoldo I, Radisic Biljak V, Bakula B, Bakula M, Simundic AM. The diagnostic accuracy of clinical and laboratory parameters in the diagnosis of acute appendicitis in the adult emergency department population - a case control pilot study. Biochem Med (Zagreb). 2018;28(3):030712.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eIkeda K, Ikeda T, Tani N, Shida A, Aoki Y, Morioka F, Ishikawa T. Acute abdominal disease with nonspecific symptoms in the elderly: A case series. J Forensic Leg Med. 2021;78:102089.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFugazzola P, Ceresoli M, Agnoletti V, Agresta F, Amato B, Carcoforo P, Catena F, Chiara O, Chiarugi M, Cobianchi L, et al. The SIFIPAC/WSES/SICG/SIMEU guidelines for diagnosis and treatment of acute appendicitis in the elderly (2019 edition). World J Emerg Surg. 2020;15(1):19.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSiddharthan RV, Byrne RM, Dewey E, Martindale RG, Gilbert EW, Tsikitis VL. Appendiceal cancer masked as inflammatory appendicitis in the elderly, not an uncommon presentation (Surveillance Epidemiology and End Results (SEER)-Medicare Analysis). J Surg Oncol. 2019;120(4):736\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBi Q, Goodman KE, Kaminsky J, Lessler J. What is Machine Learning? A Primer for the Epidemiologist. Am J Epidemiol. 2019;188(12):2222\u0026ndash;39.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBeunza JJ, Puertas E, Garc\u0026iacute;a-Ovejero E, Villalba G, Condes E, Koleva G, Hurtado C, Landecho MF. Comparison of machine learning algorithms for clinical event prediction (risk of coronary heart disease). J Biomed Inf. 2019;97:103257.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eOymaci E, Kahramansoy N, Tan S, Aydogan S, Yildirim M. The diagnostic role of preoperative blood tests in complicated appendicitis: A feasible approach to surgical decision. Niger J Clin Pract. 2023;26(7):1005\u0026ndash;10.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-surgery","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bsur","sideBox":"Learn more about [BMC Surgery](http://bmcsurg.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bsur/default.aspx","title":"BMC Surgery","twitterHandle":"@BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Machine learning, gangrenous perforation, acute appendicitis, elderly patients, diagnosis","lastPublishedDoi":"10.21203/rs.3.rs-7656285/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7656285/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e\u003cp\u003eAs life expectancy rises and elderly populations grow, acute appendicitis incidence increases, often manifesting with nonspecific symptoms that challenge diagnosis. This study applied machine learning techniques to build a predictive model for gangrenous perforation, examining clinical features and risk factors in elderly patients with acute appendicitis.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e\u003cp\u003eWe conducted a retrospective analysis of elderly patients undergoing laparoscopic appendectomy for acute appendicitis at The Second Affiliated Hospital of Kunming Medical University, China, from June 2021 to January 2024 (n\u0026thinsp;=\u0026thinsp;251). Patients were classified into gangrenous perforation (n\u0026thinsp;=\u0026thinsp;69) and non-gangrenous (n\u0026thinsp;=\u0026thinsp;182) groups, then randomly split into training (70%) and test (30%) sets. Univariate analyses, including t-tests, Spearman correlations, and chi-square tests, assessed differences across 38 variables in both sets. The least absolute shrinkage and selection operator (LASSO) screened features from the training set, informing models via logistic regression (LR), extreme gradient boosting (XGBoost), support vector machines (SVM), and random forest (RF). Model performance was evaluated using area under the receiver operating characteristic curve (AUC), with decision curve analysis assessing clinical applicability.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e\u003cp\u003eIn the training set, RF yielded the highest AUC (0.999, 95% CI: 0.998\u0026ndash;1.000), followed by XGBoost (0.975, 95% CI: 0.953\u0026ndash;0.998), LR (0.774, 95% CI: 0.692\u0026ndash;0.856), and SVM (0.768, 95% CI: 0.684\u0026ndash;0.852). In the test set, LR performed best (AUC 0.768, 95% CI: 0.642\u0026ndash;0.893), surpassing SVM (0.751, 95% CI: 0.620\u0026ndash;0.882), XGBoost (0.725, 95% CI: 0.584\u0026ndash;0.867), and RF (0.686, 95% CI: 0.533\u0026ndash;0.839). LR also showed the highest accuracy (0.784) and specificity (0.741), while XGBoost had the greatest sensitivity (0.836).\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e\u003cp\u003eAmong the models, LR emerged as the most effective for predicting gangrenous perforation in elderly acute appendicitis patients, offering robust accuracy and reliability. Its nomogram provides a noninvasive aid for clinical diagnosis.\u003c/p\u003e","manuscriptTitle":"Machine learning constructs a diagnostic prediction model for gangrenous perforation of acute appendicitis in elderly patients","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-29 04:00:30","doi":"10.21203/rs.3.rs-7656285/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-02-27T16:56:49+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-27T15:07:11+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"228486836188127867240274527735576683900","date":"2026-02-27T12:14:02+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"11052997161865660616189442533090942878","date":"2026-02-26T16:44:29+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-18T08:24:57+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-10T12:28:11+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"261609531674656751869717933161642962921","date":"2026-02-06T13:26:32+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"301762234770257001867731678634718616945","date":"2026-02-06T05:04:53+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-10-14T02:44:18+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-09-29T09:39:12+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-09-26T06:09:41+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-09-26T06:08:23+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Surgery","date":"2025-09-19T08:23:00+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-surgery","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bsur","sideBox":"Learn more about [BMC Surgery](http://bmcsurg.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bsur/default.aspx","title":"BMC Surgery","twitterHandle":"@BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"77cc030a-daf2-4e49-8015-5eb5763fe794","owner":[],"postedDate":"October 29th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2026-04-27T16:04:36+00:00","versionOfRecord":{"articleIdentity":"rs-7656285","link":"https://doi.org/10.1186/s12893-026-03753-y","journal":{"identity":"bmc-surgery","isVorOnly":false,"title":"BMC Surgery"},"publishedOn":"2026-04-24 15:58:41","publishedOnDateReadable":"April 24th, 2026"},"versionCreatedAt":"2025-10-29 04:00:30","video":"","vorDoi":"10.1186/s12893-026-03753-y","vorDoiUrl":"https://doi.org/10.1186/s12893-026-03753-y","workflowStages":[]},"version":"v1","identity":"rs-7656285","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7656285","identity":"rs-7656285","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}