Using Interpretable Machine Learning Model to Predict Orchiectomy after Testicular Torsion

Using Interpretable Machine Learning Model to Predict Orchiectomy after Testicular Torsion | 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 Using Interpretable Machine Learning Model to Predict Orchiectomy after Testicular Torsion Pengfeng Gong, Xuan Wen, Jun Zhou, Cheng Chen, Zinong Tian, You Zhao, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7692140/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 7 You are reading this latest preprint version Abstract Background This study aimed to develop and evaluate machine learning (ML) models for the preoperative prediction of orchiectomy in patients with testicular torsion. Methods We conducted a retrospective analysis of 204 cases of suspected testicular torsion managed with surgical exploration between January 2003 and April 2025. The patient cohort was partitioned via stratified sampling into a training dataset (70%) and a holdout testing dataset (30%). Initially, Least absolute shrinkage and selection operator (LASSO) regression was employed to identify six optimal predictors. Subsequently, five ML models were developed on these predictors and their performance was validated on the testing set. Model efficacy was evaluated on the testing set using the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity. Finally, we utilized Shapley Additive Explanations (SHAP) to assess model interpretability and quantify the impact of each feature. Results Among the five ML models, LightGBM model exhibited the superior predictive outcomes on the testing set. Its performance was quantified by an AUC of 0.879 (95% CI: 0.775–0.983) and an accuracy of 0.806. An examination of the model's inner workings using SHAP values determined the relative importance of each feature. The results revealed that symptom duration, degree of torsion, abdominal pain, fibrinogen, monocyte, and lymphocyte-to-monocyte ratio (LMR) were the primary drivers of its predictive power. Conclusion To predict orchiectomy in testicular torsion before surgery, we designed and validated several machine learning models. Among them, the LightGBM model using six clinical predictors demonstrated superior performance. Machine learning Orchiectomy Testicular torsion Predictive model SHAP value Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Testicular torsion refers to the twisting of the spermatic cord and testis, leading to the reduction of blood flow in testicular tissue. The most serious urological emergency accounts for 25% to 35% of acute scrotum cases in the United States 1 , 2 and Finland 3 , with a current incidence rate of approximately 3.8 per 100,000 4 . Another study indicated that 90%-100% of testicular torsion could be salvaged within 6 hours, whereas the orchiectomy rate in pediatric patients reaches 90%-100% when intervention occurs 12–24 hours after symptom onset 5 . Among pediatric patients who underwent surgery for testicular torsion, 42% ultimately required orchiectomy 4 . In the study of Goldwasser and his colleagues, the proportion of normal semen samples was only 20% after more than 12 hours of testicular torsion symptom 6 . As revealed by the previous study conducted by Arap et al, compared to 0% in the control group, 25% of patients with testicular torsion presented with oligospermia, and mean sperm counts were 38.3 million/ml, 47 million/ml and 99.3 million/ml in the orchiectomy group, the orchiopexy group and the control group, respectively 7 . Thus, identifying predictive factors for orchiectomy after testicular torsion is crucial to prevent testicular loss. According to a study from Winthrop University Hospital, degree of twisting and duration of symptoms are prognostic factors of testicular orchiectomy 8 . The study from Jefferson Einstein Healthcare Network in the United States showed that transfer status was not a predictor of testicular salvage, whereas age, time from symptom onset to presentation, and time from hospital arrival to surgery were significant predictors 9 . Predicting testicular orchiectomy or salvage by a single parameter demonstrates significant limitations. Machine learning (ML) is a branch of artificial intelligence method that enables computers to automatically learn patterns from data and make predictions or decisions 10 . ML can be used to predict cancer-specific mortality in bladder cancer patients who underwent radical cystectomy 11 . Qian Gui et.al constructed 7 ML algorithms and conventional logistic regression for predicting Gleason Grade Group upgrading (GGU) following radical prostatectomy in localized prostate cancer and concluded that logistic regression was the optimal model for predicting GGU 12 . To date, few studies have applied ML algorithms to develop predictive models for orchiectomy following testicular torsion. In this study, we collected clinical data from 204 patients with testicular torsion and developed ML-based predictive models to assist clinicians in rapidly and accurately assessing the risk of orchiectomy. Methods Patient selection We collected data from patients presenting scrotal emergencies who underwent surgical exploration in the Department of Urology at the Third Affiliated Hospital of Soochow University between January 2003 and April 2025. Patients were deemed eligible if they had an intraoperatively confirmed diagnosis of testicular torsion, supported by a preoperative scrotal ultrasound that indicated significantly reduced or absent blood flow. All included patients received either orchiectomy for testicular necrosis or detorsion with orchiopexy for a viable testis. Patients were excluded for the following reasons: (1) refusal of diagnostic surgical exploration; (2) a family request to retain a testicle diagnosed intraoperatively as necrotic; (3) incomplete medical records. Based on the surgical outcome, patients were stratified into the orchiectomy group (n = 130) and the orchidopexy group (n = 74). Ethical approval was obtained from the Medical Ethics Committee of the Third Affiliated Hospital of Soochow University (No. 2025071), and the study complied with the Declaration of Helsinki. A flowchart of the patient selection process was shown in Fig. 1 . Clinical trial number: not applicable. Data collection We retrospectively extracted patient data, which encompassed demographic details, clinical findings, and laboratory results. Demographic and clinical variables included age, body mass index (BMI), symptom duration, degree of torsion, laterality, torsion direction, abdominal pain, vomiting, scrotal tenderness, and precipitating factors. Laboratory parameters consisted of complete blood counts (leukocyte, neutrophil, lymphocyte, monocyte, platelet), derived inflammatory ratios (NLR, PLR, LMR), fibrinogen, and C-Reactive Protein (CRP). In addition, we collected healthcare pathway information, including whether the patient's first visit was to a Class A tertiary hospital and whether their first contact was with a urologist. Feature Selection Feature selection was subsequently performed with the Least Absolute Shrinkage and Selection Operator (LASSO) regression model to determine the optimal predictors. This method was selected for its ability to balance predictive accuracy with model parsimony and interpretability, which are crucial for clinical utility. Prior to model training, all continuous variables were standardized via Z-score transformation to eliminate scale-dependent biases. Through its inherent L1 regularization, LASSO penalizes and shrinks the coefficients of less influential features to zero. This process simultaneously creates a robust, streamlined set of predictors and mitigates multicollinearity (assessed via Spearman correlation) by retaining one representative from each group of highly correlated features. Model Building We first split the patient cohort into a training dataset (70%) and a testing dataset (30%). To ensure that the proportional representation of the primary outcome (orchidoexy vs. orchiectomy) was maintained in both partitions, a stratified sampling strategy was employed. Subsequently, we constructed and benchmarked five machine learning models: Logistic Regression (LR), Naive Bayes, Support Vector Machine (SVM), k-Nearest Neighbors (KNN), and Light Gradient Boosting Machine (LightGBM). All models were trained solely on the training set. The discriminative ability of each model was subsequently assessed on the testing data, using the area under the receiver operating characteristic curve (AUC) with 95% confidence intervals as the primary performance metric. Ultimately, we designated the algorithm with the highest AUC as the optimal model. Model Interpretation In this study, we placed particular emphasis on model interpretability and clinical applicability. To address the inherent “black-box” limitations of traditional machine learning algorithms, we employed the Shapley Additive Explanations (SHAP) approach. SHAP enables the quantitative assessment of each input variable’s contribution, thereby providing researchers with deeper insight into the model’s internal reasoning process. By generating a SHAP summary plot, we were able to clearly present the relative importance and global ranking of all variables, while simultaneously illustrating whether each predictor exerted a positive or negative effect on the outcome, thus enhancing the transparency and reliability of the model’s predictions. In addition, we incorporated decision curve analysis (DCA) to systematically evaluate the clinical utility of the predictive models. By plotting net benefit across different threshold probabilities, DCA provided a robust framework for determining the practical value of the model in real-world clinical scenarios. Statistical analysis Descriptive statistics were summarized according to data type. Continuous variables with non-normal distributions were reported as medians with interquartile ranges (IQR), while categorical variables were presented as absolute frequencies and percentages. Group differences were evaluated with the Mann–Whitney U test for continuous measures and the chi-squared test for categorical measures. Statistical significance was determined at a two-tailed p-value threshold of < 0.05. All analyses were conducted in Python (v3.7). Data partitioning into training and testing sets was accomplished using the StratifiedShuffleSplit method from the scikit-learn library to ensure balanced event distribution. Model construction and validation primarily leveraged the machine learning functions available in scikit-learn, whereas core statistical operations, including LASSO regression for feature selection and Spearman correlation for multicollinearity assessment, were performed with scipy, numpy, and sklearn packages. Results Patient characteristics From January 2003 to April 2025, 217 patients presenting with testicular torsion at the Third Affiliated Hospital of Soochow University were initially screened. Following the selection process, 204 patients fulfilled the inclusion criteria, of whom 130 underwent orchiectomy (Fig. 1 ). As detailed in Table 1 , significant differences were observed between the orchiectomy cohort (n = 130) and the orchidopexy cohort (n = 74) with respect to symptom duration, torsion degree, leukocyte and monocyte counts, lymphocyte-to-monocyte ratio (LMR), fibrinogen, C-reactive protein (CRP), and the presence of abdominal pain. Table 1 Baseline characteristics of patients Variables Orchidopexy (n = 74) Orchiectomy (n = 130) P value Age (years) 17 (14, 20) 15 (13, 20) 0.137 BMI (kg/m 2 ) 20.13 (18.18, 22.37) 19.72 (17.97, 22.40) 0.820 Symptoms duration (h) 6.45 (4.50, 9.28) 49.09 (25.92, 76.92) < 0.001 Degree of torsion (°) 360 (360, 540) 540 (360, 720) 0.002 Platelet (10 9 /L) 239.00 (205.50, 276.75) 239.50 (207.00, 287.50) 0.722 Leukocyte (10 9 /L) 10.26 (8.44, 11.66) 10.85 (9.31, 12.68) 0.037 Neutrophils (10 9 /L) 7.23 (5.73, 9.73) 8.10 (6.66, 9.76) 0.114 Lymphocyte (10 9 /L) 1.62 (1.16, 2.31) 1.71 (1.24, 2.38) 0.290 Monocytes (10 9 /L) 0.51 (0.36, 0.75) 0.77 (0.50, 0.94) < 0.001 NLR 4.78 (2.46, 7.34) 4.38 (3.20, 6.60) 0.767 PLR 140.72 (115.48, 189.82) 133.34 (104.45, 191.70) 0.348 LMR 2.97 (2.11, 4.10) 2.46 (1.83, 3.63) 0.043 Fibrinogen (g/L) 2.40 (1.85, 3.07) 3.10 (2.65, 3.89) < 0.001 CRP (mg/L) 13.05 (4.05, 13.05) 13.05 (7.48, 13.05) 0.021 Laterality; N (%) 1.000 Left 49(66.22) 86(66.15) Right 25(33.78) 44(33.85) Torsion direction; N (%) 0.359 Anticlockwise 50(67.57) 97(74.62) Clockwise 24(33.43) 33(25.38) First visit to a Class A tertiary hospital; N (%) 0.375 No 52(74.27) 82(63.08) Yes 22(29.73) 48(36.92) First-contact urologist; N (%) 0.214 No 50(67.57) 75(57.69) Yes 24(32.43) 55(42.31) Tender scrotum; N (%) 1.000 No 0(0.00) 0(0.00) Yes 74(100.00) 130(100.00) Abdominal pain; N (%) 0.005 No 70(94.59) 98(75.38) Yes 4(5.41) 32(24.62) Vomit; N (%) 0.768 No 73(98.65) 126(96.92) Yes 1(1.35) 4(3.08) Precipitating factor; N (%) 0.485 No 69(93.24) 116(89.23) Yes 5(6.76) 14(10.77) BMI body mass index; NLR neutrophil to lymphocyte ratio; PLR platelet to lymphocyte ratio; LMR Lymphocyte to monocyte ratio; CRP C-Reactive Protein. The dataset was subsequently divided into a training set comprising 54 orchidopexy and 88 orchiectomy patients, and a testing set with 20 orchidopexy and 42 orchiectomy patients. Baseline comparisons, along with p-values, are presented in Table 2 . Importantly, no significant difference in orchiectomy incidence was noted between the training and testing groups (p = 0.529). Furthermore, no significant disparities were identified across other baseline variables, indicating well-balanced and comparable cohorts, thus providing a robust foundation for evaluating model generalizability. Table 2 Comparison of baseline characteristics between the training set and testing set Variables Training set (n = 142) Testing set (n = 62) P value Age (years) 16 (14, 19) 17 (14, 20) 0.168 BMI (kg/m 2 ) 19.72 (18.27, 22.01) 20.03 (17.97, 23.83) 0.243 Symptoms duration (h) 26.15 (7.95, 53.64) 27.43 (7.16, 73.24) 0.453 Degree of torsion (°) 360 (360, 540) 540 (360, 720) 0.109 Platelet (10 9 /L) 240.5 (207, 283.25) 239 (207.5, 290) 0.956 Leukocyte (10 9 /L) 10.54 ± 2.8149 11.263 ± 3.0623 0.102 Neutrophil (10 9 /L) 7.705 (6.3225, 9.7) 8.245 (6.6425, 10.07) 0.301 Lymphocyte (10 9 /L) 1.65 (1.16, 2.255) 1.895 (1.4875, 2.375) 0.131 Monocyte (10 9 /L) 0.64 (0.405, 0.8675) 0.75 (0.5025, 0.97) 0.052 NLR 4.72 (2.89, 7.2775) 4.04 (3.0075, 5.95) 0.350 PLR 145.13 (110.73, 200) 130.75 (97.46, 158.73) 0.077 LMR 2.685 (1.9275, 4.02) 2.675 (2.0475, 3.585) 0.787 Fibrinogen (g/L) 3.07 (2.18, 3.59) 3.07 (2.3825, 3.6225) 0.487 CRP (mg/L) 13.05 (4.725, 13.05) 13.05 (5.475, 13.05) 0.270 Laterality; N (%) 0.999 Left 94(66.20) 41(66.13) Right 48(33.80) 21(33.87) Torsion direction; N (%) 0.536 Anticlockwise 100(70.42) 47(75.81) Clockwise 42(29.58) 15(24.19) First visit to a Class A tertiary hospital; N (%) 0.694 No 95(66.90) 39(62.90) Yes 47(33.10) 23(37.10) First-contact urologist; N (%) 0.436 No 90(63.38) 35(56.45) Yes 52(36.62) 27(43.55) Tender scrotum; N (%) 1.000 No 0(0.00) 0(0.00) Yes 142(100.00) 62(100.00) Abdominal pain; N (%) 0.823 No 118(83.10) 50(80.65) Yes 24(16.90) 12(19.35) Vomit; N (%) 0.985 No 138(97.18) 61(98.39) Yes 4(2.82) 1(1.61) Precipitating factor; N (%) 0.704 No 130(91.55) 55(88.71) Yes 12(8.45) 7(11.29) Orchiectomy; N (%) 0.529 No 54(38.03) 20(32.26) Yes 88(61.97) 42(67.74) BMI body mass index; NLR neutrophil to lymphocyte ratio; PLR platelet to lymphocyte ratio; LMR Lymphocyte to monocyte ratio; CRP C-Reactive Protein. Feature selection To identify the most representative clinical predictors, this study employed a combined strategy of Spearman correlation analysis and LASSO regression for feature selection. Spearman analysis was applied to eliminate highly correlated variables, while LASSO regression performed variable shrinkage and selection through regularization. The optimal model was obtained at a λ value of 0.0450, at which predictive performance was maximized. Following this procedure, the original 22 variables were successfully reduced to six key predictors strongly associated with orchiectomy (Fig. 2 A, B). These core predictors, as shown in Fig. 3 , were subsequently incorporated into model development and validation. Model building Five different machine learning algorithms were constructed on the basis of the six most informative variables identified during feature selection. Their predictive capability was subsequently evaluated on an independent testing set using several metrics, namely the AUC, accuracy, sensitivity, and specificity. The overall model performance for both training and testing groups is summarized in Table 3 , with ROC curves and related AUC values illustrated in Fig. 4 a. Notably, the LightGBM model showed superior discriminative ability, yielding an AUC of 0.949 (95% CI: 0.918–0.981) and accuracy of 0.852 in the training set, and maintaining robust performance in the testing set with an AUC of 0.879 (95% CI: 0.775–0.983) and accuracy of 0.806 (Fig. 4 b). Table 3 Comparison of the performance of machine learning models in the training and testing set Set Model name Accuracy AUC 95% CI Sensitivity Specificity Training set KNN 0.789 0.912 0.869–0.956 0.716 0.907 LR 0.859 0.921 0.879–0.963 0.864 0.852 NaiveBayes 0.831 0.899 0.848–0.950 0.807 0.870 SVM 0.838 0.928 0.889–0.967 0.761 0.963 LightGBM 0.852 0.949 0.918–0.981 0.795 0.944 Testing set KNN 0.581 0.752 0.621–0.884 0.452 0.850 LR 0.774 0.810 0.701–0.918 0.738 0.850 NaiveBayes 0.774 0.824 0.718–0.930 0.738 0.850 SVM 0.774 0.852 0.752–0.953 0.690 0.950 LightGBM 0.806 0.879 0.775–0.983 0.762 0.900 AUC, area under the curve; 95%CI, 95% confidence intervals; LR, logistic regression; SVM, support vector machine; KNN, K-Nearest Neighbors; LightGBM, light gradient boosting machine. Interpretation of the Optimal Model with SHAP The distribution patterns and relative contributions of potential risk factors were visualized using SHAP summary plots derived from the LightGBM model (Fig. 5 ). SHAP values enabled the quantification of each predictor’s contribution to the overall model output, thereby clarifying the importance of individual features in identifying patients at risk of orchiectomy following testicular torsion. Among all variables, symptom duration emerged as the most influential determinant, followed sequentially by fibrinogen, degree of torsion, abdominal pain, monocytes and LMR. These findings highlighted the multifactorial nature of orchiectomy risk and underscored the importance of integrating both clinical manifestations and hematological markers into predictive modeling. In addition, decision curve analysis (DCA) demonstrated a stable net clinical benefit of the LightGBM model across a broad range of threshold probabilities, supporting its potential utility in guiding clinical decision-making and optimizing patient management strategies (Fig. 6 ). Discussion Testicular torsion is one of the common emergencies in urology. According to a study from Children's Hospital of Fudan University, the orchiectomy rate was as high as 49.15% 13 , while it was 63.73% in our study. Therefore, identifying predictive factors for orchiectomy following testicular torsion has become an urgent clinical priority. A study from Norton Healthcare revealed that doppler scrotal findings, testicular and epididymal heterogeneity as well as the thickened scrotal wall, were the predictors of orchiectomy following testicular torsion 14 . In univariate logistic regression analyses, younger age, body mass index, torsion angle, red blood cells, NLR and initial presenting institution were predictive factors of testicular salvage in another Chinese study. However, in multivariate analysis, only initial presenting institution could predict testicular salvage 15 . A single predictive factor cannot effectively predict orchiectomy outcomes. Through multivariate logistic regression analysis, doctors of Shenzhen Children's Hospital identified four independent predictors of testicular salvage, consisting of symptom duration, intratesticular blood flow, degree of spermatic cord torsion and monocyte count. Based on these factors, they constructed a comprehensive nomogram for clinical prediction 16 . In this study, we constructed and compared the performance of five machine learning algorithms in predicting orchiectomy following testicular torsion. Given the inherent limitations of a retrospective single-center study, we adopted the following methodological measures: (a) standardized data quality control to minimize information bias; and (b) multivariable regression modeling to control for confounding factors, thereby enhancing inferential validity and reducing institution-specific bias. To prevent model overfitting, several strategies were implemented. First, feature selection was performed using LASSO regression, leveraging its L1 regularization property to reduce model complexity and suppress overfitting. Second, stratified five-fold cross-validation with repeated random partitioning was employed to evaluate generalization performance during model training, thus mitigating the risk of overfitting through robust out-of-sample validation. Finally, five machine learning algorithms (SVM, KNN, LR, Naive Bayes, and LightGBM) were used to construct predictive models. Comparative assessment across these algorithms based on prespecified metrics enabled the selection of the optimal model, thereby improving the generalizability of the model. Among these 5 models, the LightGBM model demonstrated the best predictive performance, achieving the highest AUC of 0.879 and the highest accuracy of 0.806. Interpretation of the LightGBM model using SHAP values revealed the relative importance and directional impact of the 6 predictors (Fig. 5 ). The predictive factors in this model consisted of symptom duration, degree of torsion, fibrinogen, monocyte, LMR and abdominal pain. The median symptom duration, which was the independent risk factor for testicular viability, was 8 hours in the orchidopexy group, and 48 hours in the orchiectomy group in a retrospective study from Anhui Provincial Children’s Hospital 17 . Another study from Brussels in Belgium demonstrated that the median prehospital time of 48 hours in the orchiectomy group was significantly longer than the 2.4 hours in the salvaged group 18 . The median symptom duration in our study was 6.45 hours and 49.09 hours in orchidopexy group and orchiectomy group, respectively. Symptom duration was identified as the most important feature for predicting the risk of orchiectomy following testicular torsion. SHAP analysis demonstrated that symptom duration was associated with high positive SHAP values, indicating that prolonged symptom duration is closely related to a significantly increased risk of orchiectomy. Testicular torsion degree refers to the angle of rotation of testis along the spermatic cord axis, which directly affects the degree of vascular obstruction and prognosis 19 . According to a study from University of Rochester School of Medicine, the median degree was 540 (range 180 to 1080) in 70 orchiectomy cases, and a median of 360 degrees (range 180 to 1080) was noted in the 116 salvage cases 19 . In the multivariate logistic regression analysis of the study from Shenzhen Children's Hospital, torsion angle was an independent factor of testicular salvage (OR = 0.995) 16 . SHAP analysis revealed that the degree of testicular torsion was associated with high positive SHAP values, reflecting a strong relationship between torsion severity and the risk of orchiectomy. Therefore, the degree of testicular torsion is also an important feature for predicting the risk of orchiectomy following testicular torsion. Although scrotal pain is the most common initial symptom of testicular torsion, some patients may also experience abdominal pain as the initial and sole symptom, which was reported in the study from Guangxi Medical University 20 . In our study, abdominal pain was a concomitant symptom of testicular pain, and it was also one of the factors predicting orchiectomy in our model. Compared to patients presenting initial testicular pain, patients with abdominal pain showed a longer symptom duration (median pain duration of 36 hours vs 5 hours) and a higher rate of testicular loss (81% vs 4%) 21 . About 12% of children may present abdominal pain as the sole initial symptom 22 , which may lead to missed scrotal examination, diagnostic delays, and an increased risk of orchiectomy 20 , 23 . Monocytes and LMR were influential predictors in predicting orchiectomy after testicular torsion in our study. A 12-year retrospective review demonstrated that monocyte count was an independent risk factors (OR = 0.02) for testicular torsion salvage in multivariate analysis 24 . The fundamental process of testicular torsion is ischemia and reperfusion injury in the testis 25 . Monocytes proliferate rapidly in the marrow and migrate to the site of testicular injury with the help of monocyte chemokine, participating in the inflammatory response 26 , 27 . The lower level of LMR in patients with peripheral arterial disease (PAD) predicts higher rates of short-term mortality and major amputations, indicating that reduced LMR levels are linked to more severe ischemic tissue damage 28 . We hypothesize that the ischemia-reperfusion injury mechanism following testicular torsion is similar to the mechanisms described above, and therefore, a decreased LMR serves as an independent predictor of orchiectomy. Fibrinogen, which acts as a coagulation factor 29 and a proinflammatory mediator 30 , emerged as an independent predictor in our model for forecasting orchiectomy, and this novel finding has never been reported in previous studies. The mechanisms of testicular ischemia-reperfusion injury primarily include hypoxia-induced energy metabolism dysfunction, calcium overload and oxidative stress initiation during the ischemic phase, while massive oxygen free radical burst and inflammatory cascade reactions during the reperfusion phase further exacerbate tissue damage, forming a vicious cycle 31 – 34 . Fibrinogen aggravates ischemia-reperfusion injury via microthrombosis and inflammatory amplification 35 . Although the mechanisms of fibrinogen in ischemia-reperfusion remain incompletely understood, we hypothesize that similar mechanisms may be involved in the testicular tissue as well, which requires further investigation to confirm. Our study excluded cases in which intraoperative diagnosis confirmed testicular necrosis, but the patients' families declined orchiectomy, to avoid the potential errors in ML models. We also applied 5 ML algorithms, and chose the optimal model (LightGBM) with the highest AUC of 0.879, avoiding algorithmic bias and improving predictive accuracy. Although we made efforts to minimize errors, some limitations are unavoidable in our retrospective study, particularly potential selection and information biases. All the patients were selected from a single hospital, and this limitation may impact the generalizability of our results. The lack of postoperative follow-up data is a limitation, as some patients who underwent testicular preservation subsequently developed testicular atrophy or functional loss. Including these cases in the testicular preservation group may compromise the model's accuracy. Additionally, the model's performance could be limited by measurement errors in symptom duration documentation, because patient recall becomes less accurate with extended symptom duration. These variables should be considered when applying this model to predict orchiectomy after testicular torsion. Conclusion We developed and validated multiple machine learning models to predict the risk of orchiectomy following testicular torsion. The LightGBM model demonstrated superior discriminative ability compared with alternative algorithms. This optimal model incorporated six key predictors, including symptom duration, degree of torsion, abdominal pain, fibrinogen, monocyte, and LMR. Implementation of this model in clinical practice may enable clinicians to assess disease progression more accurately and make timely decisions. Abbreviations ML Machine learning LASSO Least absolute shrinkage and selection operator ROC Receiver operating characteristic AUC Area under the curve SHAP Shapley additive explanations ML Machine learning LR Logistic Regression SVM Support Vector Machine KNN K-Nearest Neighbors LightGBM Light Gradient Boosting Machine NLR Neutrophil to lymphocyte ratio PLR Platelet to lymphocyte ratio LMR Lymphocyte to monocyte ratio CRP C-Reactive Protein DCA Decision curve analysis Declarations Ethics approval and consent to participate This study was conducted in accordance with the Declaration of Helsinki and was approved by the Medical Ethics Committee of the Third Affiliated Hospital of Soochow University (Approval No. 2025071). All participants provided written informed consent before participating in the study. Consent for publication Not Applicable. Availability of data and materials The datasets used and analysed during the current study are available from the corresponding author on reasonable request. Competing Interests The authors declare that they have no competing interests. Funding This research was funded by 2023 Changzhou Health Talent Overseas Training Funding Project (Municipal Health Commission of Changzhou), grant number GW2023004 (8 August 2023). Authors' contributions TZ and PG conceived and designed the study. JZ and ZT collected the data. XW and CC analyzed the data. YZ and PG made the figures and tables. TZ and PG wrote the manuscript. TZ reviewed and edited the manuscript. All authors have read and approved the final manuscript. Acknowledgements Not applicable. References Liang T, Metcalfe P, Sevcik W, Noga M. Retrospective Review of Diagnosis and Treatment in Children Presenting to the Pediatric Department With Acute Scrotum. Am J Roentgenol. 2013;200:W444–9. Waldert M, Klatte T, Schmidbauer J, Remzi M, Lackner J, Marberger M. Color Doppler sonography reliably identifies testicular torsion in boys. Urology. 2010;75:1170–4. Mäkelä E, Lahdes-Vasama T, Rajakorpi H, Wikström S. A 19-year review of paediatric patients with acute scrotum. Scand J Surg. 2007;96:62–6. Zhao LC, Lautz TB, Meeks JJ, Maizels M. Pediatric testicular torsion epidemiology using a national database: incidence, risk of orchiectomy and possible measures toward improving the quality of care. J Urol. 2011;186:2009–13. Pogorelić Z, Mustapić K, Jukić M, et al. Management of acute scrotum in children: a 25-year single center experience on 558 pediatric patients. Can J Urol. 2016;23:8594–601. Goldwasser B, Weissenberg R, Lunenfeld B, Nativ O, Many M. Semen quality and hormonal status of patients following testicular torsion. Andrologia. 1984;16:239–43. Arap MA, Vicentini FC, Cocuzza M, et al. Late hormonal levels, semen parameters, and presence of antisperm antibodies in patients treated for testicular torsion. J Androl. 2007;28:528–32. Howe AS, Vasudevan V, Kongnyuy M, et al. Degree of twisting and duration of symptoms are prognostic factors of testis salvage during episodes of testicular torsion. Translational Androl Urol. 2017;6:1159–66. Uppaluri C, Fischer K, Gaines T, et al. What matters in testicular torsion? Association of hospital transfer, race and socioeconomic factors with testicular outcomes in a single center experience. J Pediatr Urol. 2025;21:489–95. Jordan MI, Mitchell TM. Machine learning: Trends, perspectives, and prospects. Science. 2015;349:255–60. Dai L, Ye K, Yao G et al. Using machine learning for predicting cancer-specific mortality in bladder cancer patients undergoing radical cystectomy: a SEER-based study. BMC Cancer 2025;25. Gui Q, Wang X, Wu D, Guo Y. Constructing and Validating Models for Predicting Gleason Grade Group Upgrading Following Radical Prostatectomy in Localized Prostate Cancer: A Comparison between Machine Learning Algorithms and Conventional Logistic Regression. Oncology 2025:1–15. Mao C, Yang C, Zhong H et al. Development of a nomogram to predict risk factors for orchiectomy after testicular torsion in children. Sci Rep 2025;15. Shields LB, Daniels MW, Peppas DS, Rosenberg E. Sonography Findings Predict Testicular Viability in Pediatric Patients With Testicular Torsion. Cureus. 2022. Gang X-H, Duan Y-Y, Zhang B, et al. Clinical characteristics of testicular torsion and factors influencing testicular salvage in children: A 12-year study in tertiary center. World J Clin Cases. 2024;12:1251–9. Chen P, Huang W, He Y, et al. A nomogram for predicting risk factors of testicular salvage after testicular torsion in children. Int J Urol. 2024;31:568–74. Deng Q-F, Yang C, Mao C, Chu H. Clinical and hematological analysis of testicular torsion in children. Front Pead 2024;12. Steeman A, Ngatchou W, Ramadan AS, et al. Impact of treatment delays on outcome of acute testicular torsion: a 15-year retrospective study. Acta Chir Belg. 2022;122:116–22. Sessions AE, Rabinowitz R, Hulbert WC, Goldstein MM, Mevorach RA. Testicular torsion: direction, degree, duration and disinformation. J Urol. 2003;169:663–5. Wang F, Mo Z. Clinical evaluation of testicular torsion presenting with acute abdominal pain in young males. Asian J Urol. 2019;6:368–72. Vasconcelos-Castro S, Soares-Oliveira M. Abdominal pain in teenagers: Beware of testicular torsion. J Pediatr Surg. 2020;55:1933–5. Maheshwari PN, Arora AM. Let's not Clinically Miss Testicular Torsion in Patients Presenting with Lower Abdominal Pain and Vomiting. J Indian Association Pediatr Surg 2021;26. Kumar V, Matai P, Prabhu SP, Sundeep PT. Testicular Loss in Children Due to Incorrect Early Diagnosis of Torsion. Clin Pediatr (Phila). 2020;59:436–8. Chen P, Huang W, Liu L et al. Predictive value of hematological parameters in testicular salvage: A 12-year retrospective review. Front Pead 2022;10. Karaguzel E, Kadihasanoglu M, Kutlu O. Mechanisms of testicular torsion and potential protective agents. Nat Rev Urol. 2014;11:391–9. Winnall WR, Hedger MP. Phenotypic and functional heterogeneity of the testicular macrophage population: a new regulatory model. J Reprod Immunol. 2013;97:147–58. Gillette R, Tiwary R, Voss JJLP, Hewage SN, Richburg JH. Peritubular Macrophages Are Recruited to the Testis of Peripubertal Rats After Mono-(2-Ethylhexyl) Phthalate Exposure and Is Associated With Increases in the Numbers of Spermatogonia. Toxicol Sci. 2021;182:288–96. GarcÍA-Rivera E, San Norberto EM, Fidalgo-Domingos L et al. Impact of nutritional and inflammatory status in patients with critical limb-threatening ischemia. Int Angiol 2021;40. Hulshof A-M, Hemker HC, Spronk HMH, Henskens YMC, ten Cate H. Thrombin–Fibrin(ogen) Interactions, Host Defense and Risk of Thrombosis. Int J Mol Sci 2021;22. Jensen T, Kierulf P, Sandset PM, et al. Fibrinogen and fibrin induce synthesis of proinflammatory cytokines from isolated peripheral blood mononuclear cells. Thromb Haemost. 2007;97:822–9. Oyedokun PA, Akhigbe RE, Ajayi LO, Ajayi AF. Impact of hypoxia on male reproductive functions. Mol Cell Biochem. 2023;478:875–85. Akhigbe R, Ajayi A. The impact of reactive oxygen species in the development of cardiometabolic disorders: a review. Lipids Health Dis 2021;20. Nastos C, Kalimeris K, Papoutsidakis N, et al. Global Consequences of Liver Ischemia/Reperfusion Injury. Oxidative Med Cell Longev. 2014;2014:1–13. Wu H-H, Huang C-C, Chang C-P, Lin M-T, Niu K-C, Tian Y-F. Heat Shock Protein 70 (HSP70) Reduces Hepatic Inflammatory and Oxidative Damage in a Rat Model of Liver Ischemia/Reperfusion Injury with Hyperbaric Oxygen Preconditioning. Med Sci Monit. 2018;24:8096–104. Sörensen-Zender I, Rong S, Susnik N, et al. Role of fibrinogen in acute ischemic kidney injury. Am J Physiol Ren Physiol. 2013;305:F777–785. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 22 Nov, 2025 Reviewers agreed at journal 11 Nov, 2025 Reviewers invited by journal 11 Nov, 2025 Editor assigned by journal 03 Nov, 2025 Editor invited by journal 15 Oct, 2025 Submission checks completed at journal 15 Oct, 2025 First submitted to journal 15 Oct, 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-7692140","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":547146230,"identity":"32ceacba-f265-4c8e-a675-b7bfee34f637","order_by":0,"name":"Pengfeng Gong","email":"","orcid":"","institution":"The Third Affiliated Hospital of Soochow University","correspondingAuthor":false,"prefix":"","firstName":"Pengfeng","middleName":"","lastName":"Gong","suffix":""},{"id":547146231,"identity":"a8a988c9-eebc-408d-9938-95a82e6ea9ba","order_by":1,"name":"Xuan Wen","email":"","orcid":"","institution":"The Third Affiliated Hospital of Sun Yat-sen University","correspondingAuthor":false,"prefix":"","firstName":"Xuan","middleName":"","lastName":"Wen","suffix":""},{"id":547146232,"identity":"6f93a016-a007-4336-a01e-60bab50cc46c","order_by":2,"name":"Jun Zhou","email":"","orcid":"","institution":"The Third Affiliated Hospital of Soochow University","correspondingAuthor":false,"prefix":"","firstName":"Jun","middleName":"","lastName":"Zhou","suffix":""},{"id":547146233,"identity":"6b9e9270-15bb-473e-bc56-4086d0b95ece","order_by":3,"name":"Cheng Chen","email":"","orcid":"","institution":"The Third Affiliated Hospital of Soochow University","correspondingAuthor":false,"prefix":"","firstName":"Cheng","middleName":"","lastName":"Chen","suffix":""},{"id":547146234,"identity":"e717ece4-266f-41d0-a90d-80cf385ef37e","order_by":4,"name":"Zinong Tian","email":"","orcid":"","institution":"The Third Affiliated Hospital of Soochow University","correspondingAuthor":false,"prefix":"","firstName":"Zinong","middleName":"","lastName":"Tian","suffix":""},{"id":547146235,"identity":"54101039-f778-48f4-a722-16dd06a9de84","order_by":5,"name":"You Zhao","email":"","orcid":"","institution":"Liyang People's Hospital","correspondingAuthor":false,"prefix":"","firstName":"You","middleName":"","lastName":"Zhao","suffix":""},{"id":547146236,"identity":"9c5a08bf-7b2b-43c4-8e0e-6d1f156aa1ad","order_by":6,"name":"Tianwei Zhang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA/ElEQVRIiWNgGAWjYDACCQglx8/M/+CARIWEnDxxWhIYjCXbexgPWJyxMDZsIFJL4oYzZ5gPVLZVJDIcIKBDfnbzs4dffxxOnDkj98CBm/MkEhgbmB8+uoFHC+OcY+bGMgmHjfsl8hIOztwmkcfOwGZsnINHC7NEgpm0RMJt2ZkzEgwOS26TKGZs4GGTxqeFTSL9G0gL44YbQC1/50gkNhwgoIVHIsdM8kPCbUWg9w0OSDYQoUVCIqdMmiHtPzCQ2xIOSByTMDZsJuAX+Rnp2yR/2KQBo5L58AeJmjo5efbmh4/xaQEBZh5ULgHlIMD4gwhFo2AUjIJRMIIBAIoIUO5aCgIHAAAAAElFTkSuQmCC","orcid":"","institution":"The Affiliated Hospital of Qingdao University","correspondingAuthor":true,"prefix":"","firstName":"Tianwei","middleName":"","lastName":"Zhang","suffix":""}],"badges":[],"createdAt":"2025-09-23 09:08:11","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7692140/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7692140/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":96383675,"identity":"6bb1c7bc-d534-47f4-b032-01f867e2ed28","added_by":"auto","created_at":"2025-11-20 12:41:26","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2011125,"visible":true,"origin":"","legend":"","description":"","filename":"Manuscript.docx","url":"https://assets-eu.researchsquare.com/files/rs-7692140/v1/0d335e7d87d5582ac306a2a9.docx"},{"id":96383667,"identity":"45020847-1faa-4dad-8171-19bea11d959e","added_by":"auto","created_at":"2025-11-20 12:41:26","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":8216,"visible":true,"origin":"","legend":"","description":"","filename":"ce453c99abdf4b36932f7b628fd952b2.json","url":"https://assets-eu.researchsquare.com/files/rs-7692140/v1/cc9a33dc071538a1a3197e66.json"},{"id":96453435,"identity":"3e81fcc9-f800-45d7-a6c0-f97ec87f3c8b","added_by":"auto","created_at":"2025-11-21 09:59:50","extension":"xml","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":121781,"visible":true,"origin":"","legend":"","description":"","filename":"ce453c99abdf4b36932f7b628fd952b21enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7692140/v1/8bb1492ade4e7049fcd434d7.xml"},{"id":96453283,"identity":"80b6e58d-c548-4c5d-9f4a-b15c1fc9b1eb","added_by":"auto","created_at":"2025-11-21 09:59:05","extension":"jpeg","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":170642,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7692140/v1/8caa5cd28a8415c5ccc2b361.jpeg"},{"id":96383672,"identity":"63301e84-4a67-4b99-9a69-b97c8f9400a6","added_by":"auto","created_at":"2025-11-20 12:41:26","extension":"jpeg","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":500324,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7692140/v1/93e37a1457e9e4fdf2cfc1a9.jpeg"},{"id":96383674,"identity":"d4a98ee4-70f9-43aa-9cba-c44a30e4e170","added_by":"auto","created_at":"2025-11-20 12:41:26","extension":"jpeg","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":216396,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7692140/v1/dde77e11edf4658954d9e9d6.jpeg"},{"id":96383686,"identity":"843ec58e-4365-4044-a3be-b2d85a86bf20","added_by":"auto","created_at":"2025-11-20 12:41:26","extension":"jpeg","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":381506,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7692140/v1/be91ad389733bde2d572fe83.jpeg"},{"id":96383684,"identity":"a5131175-3c3c-479b-a0cb-f30234cf7114","added_by":"auto","created_at":"2025-11-20 12:41:26","extension":"jpeg","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":352588,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7692140/v1/434d77a50141088c6f551c07.jpeg"},{"id":96453887,"identity":"876323d5-4f90-4a2b-bc6e-736fd23b8dc6","added_by":"auto","created_at":"2025-11-21 10:02:02","extension":"jpeg","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":299334,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage6.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7692140/v1/3f3c76801fc81f016d65a76c.jpeg"},{"id":96453832,"identity":"8a076a3b-da10-4c89-9050-69c8afb97831","added_by":"auto","created_at":"2025-11-21 10:01:50","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":52781,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7692140/v1/685fb8890c681c24b7a37a07.png"},{"id":96383680,"identity":"41282088-e41b-4e1d-b223-2d6de648f8ff","added_by":"auto","created_at":"2025-11-20 12:41:26","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":25184,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7692140/v1/74678dd026774338ebf5a069.png"},{"id":96453526,"identity":"75e22b07-a158-40e4-bd00-2c4d889278b4","added_by":"auto","created_at":"2025-11-21 10:00:22","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":22083,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7692140/v1/a812df363582a5a637a135c1.png"},{"id":96383682,"identity":"b1ab8882-1879-4105-9163-42a48915ec02","added_by":"auto","created_at":"2025-11-20 12:41:26","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":38842,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7692140/v1/815de5b1c576d8740a53f2eb.png"},{"id":96383676,"identity":"201c088a-a5b1-4666-9398-3972ab5b7422","added_by":"auto","created_at":"2025-11-20 12:41:26","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":38153,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7692140/v1/885775096872de8d4430996f.png"},{"id":96383678,"identity":"48d2903f-9559-4fee-89ae-eebaf6fe7dfc","added_by":"auto","created_at":"2025-11-20 12:41:26","extension":"png","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":25975,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7692140/v1/adde07225f3317b73fbba12b.png"},{"id":96453294,"identity":"9783f4ae-a651-44db-a3c3-bb19e19cd082","added_by":"auto","created_at":"2025-11-21 09:59:08","extension":"xml","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":120270,"visible":true,"origin":"","legend":"","description":"","filename":"ce453c99abdf4b36932f7b628fd952b21structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7692140/v1/a5d7bd40ce626c28812d3359.xml"},{"id":96383685,"identity":"58fb9c67-358f-44ad-a1a0-683c51a73abc","added_by":"auto","created_at":"2025-11-20 12:41:26","extension":"html","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":126253,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7692140/v1/932090b5e1a410ac294de1dc.html"},{"id":96383664,"identity":"c5fd34a8-805a-418f-83ac-e9d50ee6603f","added_by":"auto","created_at":"2025-11-20 12:41:26","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":52781,"visible":true,"origin":"","legend":"\u003cp\u003eFlow chart of the process of patients’ selection\u003c/p\u003e","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7692140/v1/0cd9bfeed4bfd5d0348fc3a6.png"},{"id":96383665,"identity":"3db5cadb-42e8-45b8-b90c-ec2b3c5f60c6","added_by":"auto","created_at":"2025-11-20 12:41:26","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":25184,"visible":true,"origin":"","legend":"\u003cp\u003e(A) The process of feature selection. We used the LASSO regression model with regularization parameter (λ) tuning conducted by tenfold cross validation according to the minimum mean squared error (MSE) criteria. Based on the minimum MSE criterion, the vertical dotted line is plotted at the optimal value λ = 0.0450. (B) At this optimal λ, 6 features retain nonzero coefficients. LASSO, least absolute shrinkage and\u003c/p\u003e","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7692140/v1/caf34bdfd288486ffb25c42f.png"},{"id":96453734,"identity":"ed098436-1577-4b6a-8cae-50101dd8b81a","added_by":"auto","created_at":"2025-11-21 10:01:25","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":22083,"visible":true,"origin":"","legend":"\u003cp\u003eTop 6 selected features and the corresponding variable coefficients. Y-axis shows the top 6 variables, and X-axis shows their impact on the machine model. LMR, lymphocyte to monocyte ratio\u003c/p\u003e","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7692140/v1/952aac82ee55660dcc22e97f.png"},{"id":96383666,"identity":"e98e1b6f-60bb-427e-8f2f-8f1ec2e49141","added_by":"auto","created_at":"2025-11-20 12:41:26","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":38842,"visible":true,"origin":"","legend":"\u003cp\u003e(\u003cstrong\u003eA\u003c/strong\u003e) Performance for machine learning models in the testing set based on the AUC of the ROC curve. (\u003cstrong\u003eB\u003c/strong\u003e) AUC and the ROC curve of LightGBM model in the training set and the testing set. AUC, area under the curve; ROC, receiver operating characteristics; LR, Logistic Regression; SVM, Support Vector Machine; KNN, K-Nearest Neighbor; LightGBM, light gradient boosting machine.\u003c/p\u003e","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7692140/v1/c9823c7d7160dd872e7c3862.png"},{"id":96453037,"identity":"4a55a1b2-5e6f-4349-bdbf-9984813eee48","added_by":"auto","created_at":"2025-11-21 09:57:42","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":38153,"visible":true,"origin":"","legend":"\u003cp\u003eSHAP summary plot for the LightGBM model. Each point represents a single patient from the testing set. The y-axis lists the features, while the x-axis represents the SHAP value, indicating the impact of each feature on the model output (a positive value indicates an increased predicted risk of orchiectomy). The color of each point corresponds to the feature value for that patient, ranging from low (blue) to high (red).\u003c/p\u003e","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7692140/v1/73ea29dd8c49d3ec19ab14f7.png"},{"id":96383670,"identity":"ac4d6bdf-c2c4-408b-b8f7-9ca3451669d9","added_by":"auto","created_at":"2025-11-20 12:41:26","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":25975,"visible":true,"origin":"","legend":"\u003cp\u003eDecision curve analysis of LightGBM model in testing set.\u003c/p\u003e","description":"","filename":"Onlinefloatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7692140/v1/61c5672992aea4ac663863dd.png"},{"id":96708191,"identity":"a74e6a48-06eb-4bac-89b5-456a7361e78d","added_by":"auto","created_at":"2025-11-25 09:58:59","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1229276,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7692140/v1/6db7ee41-cfdc-4c74-9521-66b0e738f5c6.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Using Interpretable Machine Learning Model to Predict Orchiectomy after Testicular Torsion","fulltext":[{"header":"Introduction","content":"\u003cp\u003eTesticular torsion refers to the twisting of the spermatic cord and testis, leading to the reduction of blood flow in testicular tissue. The most serious urological emergency accounts for 25% to 35% of acute scrotum cases in the United States \u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e and Finland \u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e, with a current incidence rate of approximately 3.8 per 100,000 \u003csup\u003e4\u003c/sup\u003e. Another study indicated that 90%-100% of testicular torsion could be salvaged within 6 hours, whereas the orchiectomy rate in pediatric patients reaches 90%-100% when intervention occurs 12\u0026ndash;24 hours after symptom onset \u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e. Among pediatric patients who underwent surgery for testicular torsion, 42% ultimately required orchiectomy \u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e. In the study of Goldwasser and his colleagues, the proportion of normal semen samples was only 20% after more than 12 hours of testicular torsion symptom \u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e. As revealed by the previous study conducted by Arap et al, compared to 0% in the control group, 25% of patients with testicular torsion presented with oligospermia, and mean sperm counts were 38.3\u0026nbsp;million/ml, 47\u0026nbsp;million/ml and 99.3\u0026nbsp;million/ml in the orchiectomy group, the orchiopexy group and the control group, respectively \u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e. Thus, identifying predictive factors for orchiectomy after testicular torsion is crucial to prevent testicular loss. According to a study from Winthrop University Hospital, degree of twisting and duration of symptoms are prognostic factors of testicular orchiectomy \u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e. The study from Jefferson Einstein Healthcare Network in the United States showed that transfer status was not a predictor of testicular salvage, whereas age, time from symptom onset to presentation, and time from hospital arrival to surgery were significant predictors \u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e. Predicting testicular orchiectomy or salvage by a single parameter demonstrates significant limitations.\u003c/p\u003e\u003cp\u003eMachine learning (ML) is a branch of artificial intelligence method that enables computers to automatically learn patterns from data and make predictions or decisions \u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e. ML can be used to predict cancer-specific mortality in bladder cancer patients who underwent radical cystectomy \u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e. Qian Gui et.al constructed 7 ML algorithms and conventional logistic regression for predicting Gleason Grade Group upgrading (GGU) following radical prostatectomy in localized prostate cancer and concluded that logistic regression was the optimal model for predicting GGU \u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e. To date, few studies have applied ML algorithms to develop predictive models for orchiectomy following testicular torsion. In this study, we collected clinical data from 204 patients with testicular torsion and developed ML-based predictive models to assist clinicians in rapidly and accurately assessing the risk of orchiectomy.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003ePatient selection\u003c/h2\u003e\u003cp\u003eWe collected data from patients presenting scrotal emergencies who underwent surgical exploration in the Department of Urology at the Third Affiliated Hospital of Soochow University between January 2003 and April 2025. Patients were deemed eligible if they had an intraoperatively confirmed diagnosis of testicular torsion, supported by a preoperative scrotal ultrasound that indicated significantly reduced or absent blood flow. All included patients received either orchiectomy for testicular necrosis or detorsion with orchiopexy for a viable testis. Patients were excluded for the following reasons: (1) refusal of diagnostic surgical exploration; (2) a family request to retain a testicle diagnosed intraoperatively as necrotic; (3) incomplete medical records. Based on the surgical outcome, patients were stratified into the orchiectomy group (n\u0026thinsp;=\u0026thinsp;130) and the orchidopexy group (n\u0026thinsp;=\u0026thinsp;74). Ethical approval was obtained from the Medical Ethics Committee of the Third Affiliated Hospital of Soochow University (No. 2025071), and the study complied with the Declaration of Helsinki. A flowchart of the patient selection process was shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eClinical trial number: not applicable.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eData collection\u003c/h3\u003e\n\u003cp\u003eWe retrospectively extracted patient data, which encompassed demographic details, clinical findings, and laboratory results. Demographic and clinical variables included age, body mass index (BMI), symptom duration, degree of torsion, laterality, torsion direction, abdominal pain, vomiting, scrotal tenderness, and precipitating factors. Laboratory parameters consisted of complete blood counts (leukocyte, neutrophil, lymphocyte, monocyte, platelet), derived inflammatory ratios (NLR, PLR, LMR), fibrinogen, and C-Reactive Protein (CRP). In addition, we collected healthcare pathway information, including whether the patient's first visit was to a Class A tertiary hospital and whether their first contact was with a urologist.\u003c/p\u003e\n\u003ch3\u003eFeature Selection\u003c/h3\u003e\n\u003cp\u003eFeature selection was subsequently performed with the Least Absolute Shrinkage and Selection Operator (LASSO) regression model to determine the optimal predictors. This method was selected for its ability to balance predictive accuracy with model parsimony and interpretability, which are crucial for clinical utility. Prior to model training, all continuous variables were standardized via Z-score transformation to eliminate scale-dependent biases. Through its inherent L1 regularization, LASSO penalizes and shrinks the coefficients of less influential features to zero. This process simultaneously creates a robust, streamlined set of predictors and mitigates multicollinearity (assessed via Spearman correlation) by retaining one representative from each group of highly correlated features.\u003c/p\u003e\n\u003ch3\u003eModel Building\u003c/h3\u003e\n\u003cp\u003eWe first split the patient cohort into a training dataset (70%) and a testing dataset (30%). To ensure that the proportional representation of the primary outcome (orchidoexy vs. orchiectomy) was maintained in both partitions, a stratified sampling strategy was employed. Subsequently, we constructed and benchmarked five machine learning models: Logistic Regression (LR), Naive Bayes, Support Vector Machine (SVM), k-Nearest Neighbors (KNN), and Light Gradient Boosting Machine (LightGBM). All models were trained solely on the training set. The discriminative ability of each model was subsequently assessed on the testing data, using the area under the receiver operating characteristic curve (AUC) with 95% confidence intervals as the primary performance metric. Ultimately, we designated the algorithm with the highest AUC as the optimal model.\u003c/p\u003e\n\u003ch3\u003eModel Interpretation\u003c/h3\u003e\n\u003cp\u003eIn this study, we placed particular emphasis on model interpretability and clinical applicability. To address the inherent \u0026ldquo;black-box\u0026rdquo; limitations of traditional machine learning algorithms, we employed the Shapley Additive Explanations (SHAP) approach. SHAP enables the quantitative assessment of each input variable\u0026rsquo;s contribution, thereby providing researchers with deeper insight into the model\u0026rsquo;s internal reasoning process. By generating a SHAP summary plot, we were able to clearly present the relative importance and global ranking of all variables, while simultaneously illustrating whether each predictor exerted a positive or negative effect on the outcome, thus enhancing the transparency and reliability of the model\u0026rsquo;s predictions. In addition, we incorporated decision curve analysis (DCA) to systematically evaluate the clinical utility of the predictive models. By plotting net benefit across different threshold probabilities, DCA provided a robust framework for determining the practical value of the model in real-world clinical scenarios.\u003c/p\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003eStatistical analysis\u003c/h2\u003e\u003cp\u003eDescriptive statistics were summarized according to data type. Continuous variables with non-normal distributions were reported as medians with interquartile ranges (IQR), while categorical variables were presented as absolute frequencies and percentages. Group differences were evaluated with the Mann\u0026ndash;Whitney U test for continuous measures and the chi-squared test for categorical measures. Statistical significance was determined at a two-tailed p-value threshold of \u0026lt;\u0026thinsp;0.05. All analyses were conducted in Python (v3.7). Data partitioning into training and testing sets was accomplished using the StratifiedShuffleSplit method from the scikit-learn library to ensure balanced event distribution. Model construction and validation primarily leveraged the machine learning functions available in scikit-learn, whereas core statistical operations, including LASSO regression for feature selection and Spearman correlation for multicollinearity assessment, were performed with scipy, numpy, and sklearn packages.\u003c/p\u003e\u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\u003ch2\u003ePatient characteristics\u003c/h2\u003e\u003cp\u003eFrom January 2003 to April 2025, 217 patients presenting with testicular torsion at the Third Affiliated Hospital of Soochow University were initially screened. Following the selection process, 204 patients fulfilled the inclusion criteria, of whom 130 underwent orchiectomy (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). As detailed in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, significant differences were observed between the orchiectomy cohort (n\u0026thinsp;=\u0026thinsp;130) and the orchidopexy cohort (n\u0026thinsp;=\u0026thinsp;74) with respect to symptom duration, torsion degree, leukocyte and monocyte counts, lymphocyte-to-monocyte ratio (LMR), fibrinogen, C-reactive protein (CRP), and the presence of abdominal pain.\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\u003eBaseline characteristics of patients\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"4\"\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=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVariables\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eOrchidopexy (n\u0026thinsp;=\u0026thinsp;74)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eOrchiectomy (n\u0026thinsp;=\u0026thinsp;130)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\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\u003eAge\u0026nbsp;(years)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e17 (14, 20)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e15 (13, 20)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.137\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eBMI\u0026nbsp;(kg/m\u003csup\u003e2\u003c/sup\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e20.13 (18.18, 22.37)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e19.72 (17.97, 22.40)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.820\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSymptoms duration (h)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e6.45 (4.50, 9.28)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e49.09 (25.92, 76.92)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\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\u003eDegree of torsion (\u0026deg;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e360 (360, 540)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e540 (360, 720)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.002\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePlatelet (10\u003csup\u003e9\u003c/sup\u003e/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e239.00 (205.50, 276.75)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e239.50 (207.00, 287.50)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.722\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLeukocyte (10\u003csup\u003e9\u003c/sup\u003e/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e10.26 (8.44, 11.66)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e10.85 (9.31, 12.68)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.037\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNeutrophils (10\u003csup\u003e9\u003c/sup\u003e/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e7.23 (5.73, 9.73)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e8.10 (6.66, 9.76)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.114\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLymphocyte (10\u003csup\u003e9\u003c/sup\u003e/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.62 (1.16, 2.31)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1.71 (1.24, 2.38)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.290\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMonocytes (10\u003csup\u003e9\u003c/sup\u003e/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.51 (0.36, 0.75)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.77 (0.50, 0.94)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\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\u003eNLR\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e4.78 (2.46, 7.34)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e4.38 (3.20, 6.60)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.767\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePLR\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e140.72 (115.48, 189.82)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e133.34 (104.45, 191.70)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.348\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLMR\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2.97 (2.11, 4.10)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e2.46 (1.83, 3.63)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.043\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFibrinogen (g/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2.40 (1.85, 3.07)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3.10 (2.65, 3.89)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\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\u003eCRP (mg/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e13.05 (4.05, 13.05)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e13.05 (7.48, 13.05)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.021\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLaterality; N (%)\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=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.000\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLeft\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e49(66.22)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e86(66.15)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRight\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e25(33.78)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e44(33.85)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTorsion direction; N (%)\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=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.359\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAnticlockwise\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e50(67.57)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e97(74.62)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eClockwise\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e24(33.43)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e33(25.38)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFirst visit to a Class A tertiary hospital; N (%)\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=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.375\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\u003e52(74.27)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e82(63.08)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\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\u003e22(29.73)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e48(36.92)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFirst-contact urologist; N (%)\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=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.214\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\u003e50(67.57)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e75(57.69)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\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\u003e24(32.43)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e55(42.31)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTender scrotum; N (%)\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=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.000\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\u003e0(0.00)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0(0.00)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\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\u003e74(100.00)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e130(100.00)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAbdominal pain; N (%)\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=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.005\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(94.59)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e98(75.38)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\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\u003e4(5.41)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e32(24.62)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVomit; N (%)\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=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.768\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\u003e73(98.65)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e126(96.92)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\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\u003e1(1.35)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e4(3.08)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePrecipitating factor; N (%)\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=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.485\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\u003e69(93.24)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e116(89.23)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\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\u003e5(6.76)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e14(10.77)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"4\"\u003eBMI body mass index; NLR neutrophil to lymphocyte ratio; PLR platelet to lymphocyte ratio; LMR Lymphocyte to monocyte ratio; CRP C-Reactive Protein.\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eThe dataset was subsequently divided into a training set comprising 54 orchidopexy and 88 orchiectomy patients, and a testing set with 20 orchidopexy and 42 orchiectomy patients. Baseline comparisons, along with p-values, are presented in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e. Importantly, no significant difference in orchiectomy incidence was noted between the training and testing groups (p\u0026thinsp;=\u0026thinsp;0.529). Furthermore, no significant disparities were identified across other baseline variables, indicating well-balanced and comparable cohorts, thus providing a robust foundation for evaluating model generalizability.\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 of baseline characteristics between the training set and testing set\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"4\"\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=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVariables\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eTraining set (n\u0026thinsp;=\u0026thinsp;142)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eTesting set (n\u0026thinsp;=\u0026thinsp;62)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\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\u003eAge\u0026nbsp;(years)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e16 (14, 19)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e17 (14, 20)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.168\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eBMI\u0026nbsp;(kg/m\u003csup\u003e2\u003c/sup\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e19.72 (18.27, 22.01)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e20.03 (17.97, 23.83)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.243\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSymptoms duration (h)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e26.15 (7.95, 53.64)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e27.43 (7.16, 73.24)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.453\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDegree of torsion (\u0026deg;)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e360 (360, 540)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e540 (360, 720)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.109\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePlatelet (10\u003csup\u003e9\u003c/sup\u003e/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e240.5 (207, 283.25)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e239 (207.5, 290)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.956\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLeukocyte (10\u003csup\u003e9\u003c/sup\u003e/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e10.54\u0026thinsp;\u0026plusmn;\u0026thinsp;2.8149\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e11.263\u0026thinsp;\u0026plusmn;\u0026thinsp;3.0623\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.102\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNeutrophil (10\u003csup\u003e9\u003c/sup\u003e/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e7.705 (6.3225, 9.7)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e8.245 (6.6425, 10.07)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.301\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLymphocyte (10\u003csup\u003e9\u003c/sup\u003e/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.65 (1.16, 2.255)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1.895 (1.4875, 2.375)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.131\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMonocyte (10\u003csup\u003e9\u003c/sup\u003e/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.64 (0.405, 0.8675)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.75 (0.5025, 0.97)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.052\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNLR\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e4.72 (2.89, 7.2775)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e4.04 (3.0075, 5.95)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.350\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePLR\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e145.13 (110.73, 200)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e130.75 (97.46, 158.73)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.077\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLMR\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2.685 (1.9275, 4.02)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e2.675 (2.0475, 3.585)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.787\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFibrinogen (g/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e3.07 (2.18, 3.59)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3.07 (2.3825, 3.6225)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.487\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCRP (mg/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e13.05 (4.725, 13.05)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e13.05 (5.475, 13.05)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.270\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLaterality; N (%)\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=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.999\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLeft\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e94(66.20)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e41(66.13)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRight\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e48(33.80)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e21(33.87)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTorsion direction; N (%)\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=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.536\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAnticlockwise\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e100(70.42)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e47(75.81)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eClockwise\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e42(29.58)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e15(24.19)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFirst visit to a Class A tertiary hospital; N (%)\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=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.694\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\u003e95(66.90)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e39(62.90)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\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\u003e47(33.10)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e23(37.10)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFirst-contact urologist; N (%)\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=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.436\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\u003e90(63.38)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e35(56.45)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\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\u003e52(36.62)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e27(43.55)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTender scrotum; N (%)\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=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.000\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\u003e0(0.00)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0(0.00)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\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\u003e142(100.00)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e62(100.00)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAbdominal pain; N (%)\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=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.823\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\u003e118(83.10)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e50(80.65)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\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\u003e24(16.90)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e12(19.35)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVomit; N (%)\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=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.985\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\u003e138(97.18)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e61(98.39)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\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\u003e4(2.82)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1(1.61)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePrecipitating factor; N (%)\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=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.704\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\u003e130(91.55)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e55(88.71)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\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\u003e12(8.45)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e7(11.29)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eOrchiectomy; N (%)\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=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.529\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\u003e54(38.03)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e20(32.26)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\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\u003e88(61.97)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e42(67.74)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"4\"\u003eBMI body mass index; NLR neutrophil to lymphocyte ratio; PLR platelet to lymphocyte ratio; LMR Lymphocyte to monocyte ratio; CRP C-Reactive Protein.\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003eFeature selection\u003c/h2\u003e\u003cp\u003eTo identify the most representative clinical predictors, this study employed a combined strategy of Spearman correlation analysis and LASSO regression for feature selection. Spearman analysis was applied to eliminate highly correlated variables, while LASSO regression performed variable shrinkage and selection through regularization. The optimal model was obtained at a λ value of 0.0450, at which predictive performance was maximized. Following this procedure, the original 22 variables were successfully reduced to six key predictors strongly associated with orchiectomy (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA, B). These core predictors, as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, were subsequently incorporated into model development and validation.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003eModel building\u003c/h2\u003e\u003cp\u003eFive different machine learning algorithms were constructed on the basis of the six most informative variables identified during feature selection. Their predictive capability was subsequently evaluated on an independent testing set using several metrics, namely the AUC, accuracy, sensitivity, and specificity. The overall model performance for both training and testing groups is summarized in Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, with ROC curves and related AUC values illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea. Notably, the LightGBM model showed superior discriminative ability, yielding an AUC of 0.949 (95% CI: 0.918\u0026ndash;0.981) and accuracy of 0.852 in the training set, and maintaining robust performance in the testing set with an AUC of 0.879 (95% CI: 0.775\u0026ndash;0.983) and accuracy of 0.806 (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eb).\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 of the performance of machine learning models in the training and testing set\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=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSet\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eModel name\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\u003eAUC\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003e95% CI\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eSensitivity\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u003cp\u003eSpecificity\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTraining set\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eKNN\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.789\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.912\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.869\u0026ndash;0.956\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.716\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.907\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\u003eLR\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.859\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.921\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.879\u0026ndash;0.963\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.864\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.852\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\u003eNaiveBayes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.831\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.899\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.848\u0026ndash;0.950\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.807\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.870\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\u003eSVM\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.838\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.928\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.889\u0026ndash;0.967\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.761\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.963\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\u003eLightGBM\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.852\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.949\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.918\u0026ndash;0.981\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.795\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.944\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTesting set\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eKNN\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.581\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.752\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.621\u0026ndash;0.884\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.452\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.850\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\u003eLR\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.774\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.810\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.701\u0026ndash;0.918\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.738\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.850\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\u003eNaiveBayes\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.774\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.824\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.718\u0026ndash;0.930\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.738\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.850\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\u003eSVM\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.774\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.852\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.752\u0026ndash;0.953\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.690\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.950\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\u003eLightGBM\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.806\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.879\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.775\u0026ndash;0.983\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.762\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.900\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"7\"\u003eAUC, area under the curve; 95%CI, 95% confidence intervals; LR, logistic regression; SVM, support vector machine; KNN, K-Nearest Neighbors; LightGBM, light gradient boosting machine.\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\u003ch2\u003eInterpretation of the Optimal Model with SHAP\u003c/h2\u003e\u003cp\u003eThe distribution patterns and relative contributions of potential risk factors were visualized using SHAP summary plots derived from the LightGBM model (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). SHAP values enabled the quantification of each predictor\u0026rsquo;s contribution to the overall model output, thereby clarifying the importance of individual features in identifying patients at risk of orchiectomy following testicular torsion. Among all variables, symptom duration emerged as the most influential determinant, followed sequentially by fibrinogen, degree of torsion, abdominal pain, monocytes and LMR. These findings highlighted the multifactorial nature of orchiectomy risk and underscored the importance of integrating both clinical manifestations and hematological markers into predictive modeling. In addition, decision curve analysis (DCA) demonstrated a stable net clinical benefit of the LightGBM model across a broad range of threshold probabilities, supporting its potential utility in guiding clinical decision-making and optimizing patient management strategies (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eTesticular torsion is one of the common emergencies in urology. According to a study from Children's Hospital of Fudan University, the orchiectomy rate was as high as 49.15% \u003csup\u003e13\u003c/sup\u003e, while it was 63.73% in our study. Therefore, identifying predictive factors for orchiectomy following testicular torsion has become an urgent clinical priority. A study from Norton Healthcare revealed that doppler scrotal findings, testicular and epididymal heterogeneity as well as the thickened scrotal wall, were the predictors of orchiectomy following testicular torsion \u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e. In univariate logistic regression analyses, younger age, body mass index, torsion angle, red blood cells, NLR and initial presenting institution were predictive factors of testicular salvage in another Chinese study. However, in multivariate analysis, only initial presenting institution could predict testicular salvage \u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e. A single predictive factor cannot effectively predict orchiectomy outcomes. Through multivariate logistic regression analysis, doctors of Shenzhen Children's Hospital identified four independent predictors of testicular salvage, consisting of symptom duration, intratesticular blood flow, degree of spermatic cord torsion and monocyte count. Based on these factors, they constructed a comprehensive nomogram for clinical prediction \u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eIn this study, we constructed and compared the performance of five machine learning algorithms in predicting orchiectomy following testicular torsion. Given the inherent limitations of a retrospective single-center study, we adopted the following methodological measures: (a) standardized data quality control to minimize information bias; and (b) multivariable regression modeling to control for confounding factors, thereby enhancing inferential validity and reducing institution-specific bias. To prevent model overfitting, several strategies were implemented. First, feature selection was performed using LASSO regression, leveraging its L1 regularization property to reduce model complexity and suppress overfitting. Second, stratified five-fold cross-validation with repeated random partitioning was employed to evaluate generalization performance during model training, thus mitigating the risk of overfitting through robust out-of-sample validation. Finally, five machine learning algorithms (SVM, KNN, LR, Naive Bayes, and LightGBM) were used to construct predictive models. Comparative assessment across these algorithms based on prespecified metrics enabled the selection of the optimal model, thereby improving the generalizability of the model.\u003c/p\u003e\u003cp\u003eAmong these 5 models, the LightGBM model demonstrated the best predictive performance, achieving the highest AUC of 0.879 and the highest accuracy of 0.806. Interpretation of the LightGBM model using SHAP values revealed the relative importance and directional impact of the 6 predictors (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). The predictive factors in this model consisted of symptom duration, degree of torsion, fibrinogen, monocyte, LMR and abdominal pain.\u003c/p\u003e\u003cp\u003eThe median symptom duration, which was the independent risk factor for testicular viability, was 8 hours in the orchidopexy group, and 48 hours in the orchiectomy group in a retrospective study from Anhui Provincial Children\u0026rsquo;s Hospital \u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e. Another study from Brussels in Belgium demonstrated that the median prehospital time of 48 hours in the orchiectomy group was significantly longer than the 2.4 hours in the salvaged group \u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e. The median symptom duration in our study was 6.45 hours and 49.09 hours in orchidopexy group and orchiectomy group, respectively. Symptom duration was identified as the most important feature for predicting the risk of orchiectomy following testicular torsion. SHAP analysis demonstrated that symptom duration was associated with high positive SHAP values, indicating that prolonged symptom duration is closely related to a significantly increased risk of orchiectomy.\u003c/p\u003e\u003cp\u003eTesticular torsion degree refers to the angle of rotation of testis along the spermatic cord axis, which directly affects the degree of vascular obstruction and prognosis \u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e. According to a study from University of Rochester School of Medicine, the median degree was 540 (range 180 to 1080) in 70 orchiectomy cases, and a median of 360 degrees (range 180 to 1080) was noted in the 116 salvage cases \u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e. In the multivariate logistic regression analysis of the study from Shenzhen Children's Hospital, torsion angle was an independent factor of testicular salvage (OR\u0026thinsp;=\u0026thinsp;0.995) \u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e. SHAP analysis revealed that the degree of testicular torsion was associated with high positive SHAP values, reflecting a strong relationship between torsion severity and the risk of orchiectomy. Therefore, the degree of testicular torsion is also an important feature for predicting the risk of orchiectomy following testicular torsion.\u003c/p\u003e\u003cp\u003eAlthough scrotal pain is the most common initial symptom of testicular torsion, some patients may also experience abdominal pain as the initial and sole symptom, which was reported in the study from Guangxi Medical University \u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e. In our study, abdominal pain was a concomitant symptom of testicular pain, and it was also one of the factors predicting orchiectomy in our model. Compared to patients presenting initial testicular pain, patients with abdominal pain showed a longer symptom duration (median pain duration of 36 hours vs 5 hours) and a higher rate of testicular loss (81% vs 4%) \u003csup\u003e21\u003c/sup\u003e. About 12% of children may present abdominal pain as the sole initial symptom \u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e, which may lead to missed scrotal examination, diagnostic delays, and an increased risk of orchiectomy \u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eMonocytes and LMR were influential predictors in predicting orchiectomy after testicular torsion in our study. A 12-year retrospective review demonstrated that monocyte count was an independent risk factors (OR\u0026thinsp;=\u0026thinsp;0.02) for testicular torsion salvage in multivariate analysis \u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e. The fundamental process of testicular torsion is ischemia and reperfusion injury in the testis \u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e. Monocytes proliferate rapidly in the marrow and migrate to the site of testicular injury with the help of monocyte chemokine, participating in the inflammatory response \u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e. The lower level of LMR in patients with peripheral arterial disease (PAD) predicts higher rates of short-term mortality and major amputations, indicating that reduced LMR levels are linked to more severe ischemic tissue damage \u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e. We hypothesize that the ischemia-reperfusion injury mechanism following testicular torsion is similar to the mechanisms described above, and therefore, a decreased LMR serves as an independent predictor of orchiectomy.\u003c/p\u003e\u003cp\u003eFibrinogen, which acts as a coagulation factor \u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e and a proinflammatory mediator \u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e, emerged as an independent predictor in our model for forecasting orchiectomy, and this novel finding has never been reported in previous studies. The mechanisms of testicular ischemia-reperfusion injury primarily include hypoxia-induced energy metabolism dysfunction, calcium overload and oxidative stress initiation during the ischemic phase, while massive oxygen free radical burst and inflammatory cascade reactions during the reperfusion phase further exacerbate tissue damage, forming a vicious cycle \u003csup\u003e\u003cspan additionalcitationids=\"CR32 CR33\" citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e. Fibrinogen aggravates ischemia-reperfusion injury via microthrombosis and inflammatory amplification \u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e. Although the mechanisms of fibrinogen in ischemia-reperfusion remain incompletely understood, we hypothesize that similar mechanisms may be involved in the testicular tissue as well, which requires further investigation to confirm.\u003c/p\u003e\u003cp\u003eOur study excluded cases in which intraoperative diagnosis confirmed testicular necrosis, but the patients' families declined orchiectomy, to avoid the potential errors in ML models. We also applied 5 ML algorithms, and chose the optimal model (LightGBM) with the highest AUC of 0.879, avoiding algorithmic bias and improving predictive accuracy. Although we made efforts to minimize errors, some limitations are unavoidable in our retrospective study, particularly potential selection and information biases. All the patients were selected from a single hospital, and this limitation may impact the generalizability of our results. The lack of postoperative follow-up data is a limitation, as some patients who underwent testicular preservation subsequently developed testicular atrophy or functional loss. Including these cases in the testicular preservation group may compromise the model's accuracy. Additionally, the model's performance could be limited by measurement errors in symptom duration documentation, because patient recall becomes less accurate with extended symptom duration. These variables should be considered when applying this model to predict orchiectomy after testicular torsion.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eWe developed and validated multiple machine learning models to predict the risk of orchiectomy following testicular torsion. The LightGBM model demonstrated superior discriminative ability compared with alternative algorithms. This optimal model incorporated six key predictors, including symptom duration, degree of torsion, abdominal pain, fibrinogen, monocyte, and LMR. Implementation of this model in clinical practice may enable clinicians to assess disease progression more accurately and make timely decisions.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 170px;\"\u003e\n \u003cp\u003eML\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 383px;\"\u003e\n \u003cp\u003eMachine learning\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 170px;\"\u003e\n \u003cp\u003eLASSO\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 383px;\"\u003e\n \u003cp\u003eLeast absolute shrinkage and selection operator\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 170px;\"\u003e\n \u003cp\u003eROC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 383px;\"\u003e\n \u003cp\u003eReceiver operating characteristic\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 170px;\"\u003e\n \u003cp\u003eAUC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 383px;\"\u003e\n \u003cp\u003eArea under the curve\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 170px;\"\u003e\n \u003cp\u003eSHAP\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 383px;\"\u003e\n \u003cp\u003eShapley additive explanations\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 170px;\"\u003e\n \u003cp\u003eML\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 383px;\"\u003e\n \u003cp\u003eMachine learning\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003eLR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 383px;\"\u003e\n \u003cp\u003eLogistic Regression\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003eSVM\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 383px;\"\u003e\n \u003cp\u003eSupport Vector Machine\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003eKNN\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 383px;\"\u003e\n \u003cp\u003eK-Nearest Neighbors\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003eLightGBM\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 383px;\"\u003e\n \u003cp\u003eLight Gradient Boosting Machine\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003eNLR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 383px;\"\u003e\n \u003cp\u003eNeutrophil to lymphocyte ratio\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003ePLR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 383px;\"\u003e\n \u003cp\u003ePlatelet to lymphocyte ratio\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003eLMR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 383px;\"\u003e\n \u003cp\u003eLymphocyte to monocyte ratio\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003eCRP\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 383px;\"\u003e\n \u003cp\u003eC-Reactive Protein\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003eDCA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 383px;\"\u003e\n \u003cp\u003eDecision curve analysis\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was conducted in accordance with the Declaration of Helsinki and was approved by the Medical Ethics Committee of the Third Affiliated Hospital of Soochow University (Approval No. 2025071). All participants provided written informed consent before participating in the study.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot Applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets used and analysed during the current study are available from the corresponding author on reasonable request.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting Interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis research was funded by 2023 Changzhou Health Talent Overseas Training Funding Project (Municipal Health Commission of Changzhou), grant number GW2023004 (8 August 2023).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026apos; contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTZ and PG conceived and designed the study. JZ and ZT collected the data. XW and CC analyzed the data. YZ and PG made the figures and tables. TZ and PG wrote the manuscript. TZ reviewed and edited the manuscript. All authors have read and approved the final manuscript.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eLiang T, Metcalfe P, Sevcik W, Noga M. Retrospective Review of Diagnosis and Treatment in Children Presenting to the Pediatric Department With Acute Scrotum. Am J Roentgenol. 2013;200:W444\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWaldert M, Klatte T, Schmidbauer J, Remzi M, Lackner J, Marberger M. Color Doppler sonography reliably identifies testicular torsion in boys. Urology. 2010;75:1170\u0026ndash;4.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eM\u0026auml;kel\u0026auml; E, Lahdes-Vasama T, Rajakorpi H, Wikstr\u0026ouml;m S. A 19-year review of paediatric patients with acute scrotum. Scand J Surg. 2007;96:62\u0026ndash;6.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhao LC, Lautz TB, Meeks JJ, Maizels M. Pediatric testicular torsion epidemiology using a national database: incidence, risk of orchiectomy and possible measures toward improving the quality of care. J Urol. 2011;186:2009\u0026ndash;13.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePogorelić Z, Mustapić K, Jukić M, et al. Management of acute scrotum in children: a 25-year single center experience on 558 pediatric patients. Can J Urol. 2016;23:8594\u0026ndash;601.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGoldwasser B, Weissenberg R, Lunenfeld B, Nativ O, Many M. Semen quality and hormonal status of patients following testicular torsion. Andrologia. 1984;16:239\u0026ndash;43.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eArap MA, Vicentini FC, Cocuzza M, et al. Late hormonal levels, semen parameters, and presence of antisperm antibodies in patients treated for testicular torsion. J Androl. 2007;28:528\u0026ndash;32.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHowe AS, Vasudevan V, Kongnyuy M, et al. Degree of twisting and duration of symptoms are prognostic factors of testis salvage during episodes of testicular torsion. Translational Androl Urol. 2017;6:1159\u0026ndash;66.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eUppaluri C, Fischer K, Gaines T, et al. What matters in testicular torsion? Association of hospital transfer, race and socioeconomic factors with testicular outcomes in a single center experience. J Pediatr Urol. 2025;21:489\u0026ndash;95.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJordan MI, Mitchell TM. Machine learning: Trends, perspectives, and prospects. Science. 2015;349:255\u0026ndash;60.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDai L, Ye K, Yao G et al. Using machine learning for predicting cancer-specific mortality in bladder cancer patients undergoing radical cystectomy: a SEER-based study. BMC Cancer 2025;25.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGui Q, Wang X, Wu D, Guo Y. Constructing and Validating Models for Predicting Gleason Grade Group Upgrading Following Radical Prostatectomy in Localized Prostate Cancer: A Comparison between Machine Learning Algorithms and Conventional Logistic Regression. Oncology 2025:1\u0026ndash;15.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMao C, Yang C, Zhong H et al. Development of a nomogram to predict risk factors for orchiectomy after testicular torsion in children. Sci Rep 2025;15.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eShields LB, Daniels MW, Peppas DS, Rosenberg E. Sonography Findings Predict Testicular Viability in Pediatric Patients With Testicular Torsion. Cureus. 2022.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGang X-H, Duan Y-Y, Zhang B, et al. Clinical characteristics of testicular torsion and factors influencing testicular salvage in children: A 12-year study in tertiary center. World J Clin Cases. 2024;12:1251\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChen P, Huang W, He Y, et al. A nomogram for predicting risk factors of testicular salvage after testicular torsion in children. Int J Urol. 2024;31:568\u0026ndash;74.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDeng Q-F, Yang C, Mao C, Chu H. Clinical and hematological analysis of testicular torsion in children. Front Pead 2024;12.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSteeman A, Ngatchou W, Ramadan AS, et al. Impact of treatment delays on outcome of acute testicular torsion: a 15-year retrospective study. Acta Chir Belg. 2022;122:116\u0026ndash;22.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSessions AE, Rabinowitz R, Hulbert WC, Goldstein MM, Mevorach RA. Testicular torsion: direction, degree, duration and disinformation. J Urol. 2003;169:663\u0026ndash;5.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang F, Mo Z. Clinical evaluation of testicular torsion presenting with acute abdominal pain in young males. Asian J Urol. 2019;6:368\u0026ndash;72.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVasconcelos-Castro S, Soares-Oliveira M. Abdominal pain in teenagers: Beware of testicular torsion. J Pediatr Surg. 2020;55:1933\u0026ndash;5.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMaheshwari PN, Arora AM. Let's not Clinically Miss Testicular Torsion in Patients Presenting with Lower Abdominal Pain and Vomiting. J Indian Association Pediatr Surg 2021;26.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKumar V, Matai P, Prabhu SP, Sundeep PT. Testicular Loss in Children Due to Incorrect Early Diagnosis of Torsion. Clin Pediatr (Phila). 2020;59:436\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChen P, Huang W, Liu L et al. Predictive value of hematological parameters in testicular salvage: A 12-year retrospective review. Front Pead 2022;10.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKaraguzel E, Kadihasanoglu M, Kutlu O. Mechanisms of testicular torsion and potential protective agents. Nat Rev Urol. 2014;11:391\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWinnall WR, Hedger MP. Phenotypic and functional heterogeneity of the testicular macrophage population: a new regulatory model. J Reprod Immunol. 2013;97:147\u0026ndash;58.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGillette R, Tiwary R, Voss JJLP, Hewage SN, Richburg JH. Peritubular Macrophages Are Recruited to the Testis of Peripubertal Rats After Mono-(2-Ethylhexyl) Phthalate Exposure and Is Associated With Increases in the Numbers of Spermatogonia. Toxicol Sci. 2021;182:288\u0026ndash;96.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGarc\u0026Iacute;A-Rivera E, San Norberto EM, Fidalgo-Domingos L et al. Impact of nutritional and inflammatory status in patients with critical limb-threatening ischemia. Int Angiol 2021;40.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHulshof A-M, Hemker HC, Spronk HMH, Henskens YMC, ten Cate H. Thrombin\u0026ndash;Fibrin(ogen) Interactions, Host Defense and Risk of Thrombosis. Int J Mol Sci 2021;22.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJensen T, Kierulf P, Sandset PM, et al. Fibrinogen and fibrin induce synthesis of proinflammatory cytokines from isolated peripheral blood mononuclear cells. Thromb Haemost. 2007;97:822\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eOyedokun PA, Akhigbe RE, Ajayi LO, Ajayi AF. Impact of hypoxia on male reproductive functions. Mol Cell Biochem. 2023;478:875\u0026ndash;85.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAkhigbe R, Ajayi A. The impact of reactive oxygen species in the development of cardiometabolic disorders: a review. Lipids Health Dis 2021;20.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNastos C, Kalimeris K, Papoutsidakis N, et al. Global Consequences of Liver Ischemia/Reperfusion Injury. Oxidative Med Cell Longev. 2014;2014:1\u0026ndash;13.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWu H-H, Huang C-C, Chang C-P, Lin M-T, Niu K-C, Tian Y-F. Heat Shock Protein 70 (HSP70) Reduces Hepatic Inflammatory and Oxidative Damage in a Rat Model of Liver Ischemia/Reperfusion Injury with Hyperbaric Oxygen Preconditioning. Med Sci Monit. 2018;24:8096\u0026ndash;104.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eS\u0026ouml;rensen-Zender I, Rong S, Susnik N, et al. Role of fibrinogen in acute ischemic kidney injury. Am J Physiol Ren Physiol. 2013;305:F777\u0026ndash;785.\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":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-urology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"buro","sideBox":"Learn more about [BMC Urology](http://bmcurol.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/buro/default.aspx","title":"BMC Urology","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Machine learning, Orchiectomy, Testicular torsion, Predictive model, SHAP value","lastPublishedDoi":"10.21203/rs.3.rs-7692140/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7692140/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e\u003cp\u003eThis study aimed to develop and evaluate machine learning (ML) models for the preoperative prediction of orchiectomy in patients with testicular torsion.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e\u003cp\u003eWe conducted a retrospective analysis of 204 cases of suspected testicular torsion managed with surgical exploration between January 2003 and April 2025. The patient cohort was partitioned via stratified sampling into a training dataset (70%) and a holdout testing dataset (30%). Initially, Least absolute shrinkage and selection operator (LASSO) regression was employed to identify six optimal predictors. Subsequently, five ML models were developed on these predictors and their performance was validated on the testing set. Model efficacy was evaluated on the testing set using the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity. Finally, we utilized Shapley Additive Explanations (SHAP) to assess model interpretability and quantify the impact of each feature.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e\u003cp\u003eAmong the five ML models, LightGBM model exhibited the superior predictive outcomes on the testing set. Its performance was quantified by an AUC of 0.879 (95% CI: 0.775\u0026ndash;0.983) and an accuracy of 0.806. An examination of the model's inner workings using SHAP values determined the relative importance of each feature. The results revealed that symptom duration, degree of torsion, abdominal pain, fibrinogen, monocyte, and lymphocyte-to-monocyte ratio (LMR) were the primary drivers of its predictive power.\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e\u003cp\u003eTo predict orchiectomy in testicular torsion before surgery, we designed and validated several machine learning models. Among them, the LightGBM model using six clinical predictors demonstrated superior performance.\u003c/p\u003e","manuscriptTitle":"Using Interpretable Machine Learning Model to Predict Orchiectomy after Testicular Torsion","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-20 12:41:21","doi":"10.21203/rs.3.rs-7692140/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2025-11-22T11:23:56+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"87308281322048996032439108840805402963","date":"2025-11-11T06:11:31+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-11-11T05:52:38+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-11-03T12:08:44+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-10-15T13:46:12+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-10-15T09:16:37+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Urology","date":"2025-10-15T09:13:51+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-urology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"buro","sideBox":"Learn more about [BMC Urology](http://bmcurol.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/buro/default.aspx","title":"BMC Urology","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"2e4aea50-2b6b-44c0-a81e-02b74896085f","owner":[],"postedDate":"November 20th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2025-11-20T12:41:21+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-20 12:41:21","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7692140","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7692140","identity":"rs-7692140","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}