Prediction and evaluation of the risk of Stroke-associated pneumonia using an artificial neural network model

Prediction and evaluation of the risk of Stroke-associated pneumonia using an artificial neural network model | 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 Prediction and evaluation of the risk of Stroke-associated pneumonia using an artificial neural network model Ting Wang, Chunbiao Li, Linli Yuan, Min You, Juan Yuan, Sijing Peng, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4754561/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Objective This study developed a predictive model for the risk of stroke-associated pneumonia (SAP) based on an advanced artificial neural network (ANN) model. Methods Using clinical and laboratory data collected from a cohort of 456 hospital patients from July 2017 to April 2022, we constructed ANN and logistic regression (LR) models. The models were trained on a randomly selected group of 292 patients, and subsequent model validation and testing were carried out on two separate sets of 82 patients each. The predictive performances of both models were evaluated using a comprehensive range of statistical indices. Results During dataset partitioning, the 24 variables across the training, validation, and test sets displayed no significant discrepancies. The prediction performance of the ANN model was better than that of the LR model. When applied to the test cohort, the ANN model had a sensitivity of 83.53% and a specificity of 85.18%. Comparative analysis revealed discernible discrepancies between the performance indexes of the ANN and LR models. Based on the receiver operating characteristic curve, the ANN model showed robust ability to identify SAP, with an area under the curve value of 0.920. The principal independent predictors in the model were serum albumin, activities of daily living score, hemoglobin level, and hypersensitive C-reactive protein level. Conclusions The developed ANN model demonstrates promising predictive capability for assessing the risk of SAP. However, further verification with larger and more diverse datasets is needed to confirm its utility as a tool for clinical prediction. Stroke Stroke-associated pneumonia Artificial neural networks risk prediction Figures Figure 1 Figure 2 Figure 3 Figure 4 Introduction Stroke-associated pneumonia (SAP), which occurs within seven days of onset in non-ventilated stroke patients, is one of the most common complications of cerebrovascular accidents. [ 1 ] Surveys indicate that between 5% and 65% of stroke patients develop SAP, [ 2 , 3 ] which can increase the mortality rate, prolong the hospital stay, and impose a significant economic burden on the patient. [ 4 , 5 ] As a result, the early identification and assessment of SAP risk is critical for improving patient prognosis and reducing healthcare costs. Numerous risk prediction models with varying degrees of predictive effectiveness have been developed for predicting the occurrence of SAP. [ 6 – 10 ] However, these models fail to consider the comprehensive range of risk factors and associated variables. Furthermore, the accuracy and stability of typical statistical methods such as logistic regression (LR) are compromised when dealing with nonlinear relationships and high-dimensional interactions. Artificial neural networks (ANNs), also termed multi-layer perceptrons, are deep neural networks that can be used to model problems in which the relationships between causal factors and responses are complex. [ 11 ] In recent years, ANNs have been increasingly used as powerful machine learning tools in risk prediction models. [ 12 – 14 ] ANNs are capable of learning and simulating complex nonlinear relationships between variables; more importantly, they have the ability to self-learn, which allows them to optimize the model by progressively adjusting the weights within the network. Consequently, the resulting ANNs may demonstrate superior performance compared with traditional statistical models when identifying risk factors for SAP. [ 15 – 17 ] In this study, we developed an ANN model to predict the risk of SAP in stroke patients and compared its predictive performance with that of a traditional LR model. Our goal was to construct an accurate prediction tool that can effectively identify patients at high risk of SAP, thereby providing support for clinical decision-making. Methods General information All stroke patients were enrolled in the clinical departments of our hospital from July 2017 to April 2022. The inclusion criteria were as follows: (1) age ≥ 18; and (2) stroke diagnosis was confirmed via computed tomography or magnetic resonance imaging of the skull. The exclusion criteria were as follows: (1) patients with pre-existing pneumonia prior to admission; and (2) a history of hematological diseases, malignancy, immunosuppressant treatment, or severe hepatic or renal dysfunction. SAP was diagnosed based on clinical, laboratory, and radiological data by the treating physician and recorded on the medical chart. International guidelines were followed by all the patients in terms of standard medical treatment. A total of 456 patients (280 men and 176 women) participated in this study. The protocol for the study was formulated in accordance with the guidelines of the institutional ethical committee. Informed consent was given by all participants, and the research was carried out in accordance with the ethical principles outlined in the Helsinki Declaration. After conducting a thorough examination of literature, we selected 24 factors to incorporate into the ANN model: demographic factors (gender and age); medical history (chronic obstructive pulmonary disease, high blood pressure, diabetes, atrial fibrillation, smoking history, and drinking history); laboratory data (white blood cell count, neutrophil count, lymphocyte count, hypersensitive C-reactive protein [CRP] level, albumin level, triglyceride level, and hemoglobin level); and hospitalization characteristics (activities of daily living score, stroke quality, stroke position, disorders of consciousness, and dysphagia). ANN model The ANN model consisted of three parts: an input layer, a hidden layer, and an output layer(Fig. 1 ). [ 18 ] The input layer was a feed-forward multilayer structure consisting of nonlinear neurons (i.e., perceptrons), and every layer consisted of anthropomorphic neurons modeled by nodes. [ 19 ] The ANN model used an integrated set of 24 input variables; Table 1 gives an overview of these input variables along with their means, standard deviations, and frequencies. If the inputs to the input hierarchy are set to x(1), x(2),..., x(R), the output of the input layer corresponding to the input of the hidden layer neurons can be represented as \(\:s\left(i\right)={\sum\:}_{j=1}^{R1}W\left(i,j\right)\text{*}x\left(j\right)-b1\left(i\right);\:y\left(i\right)=\text{f}\left(s\right(i\left)\right),i=\text{1,2},...,R2,\) where W ( i , j ) denotes the connection weight of the input layer neurons ( i ) and the hidden layer neurons ( j ), and b 1( j ) represents the threshold of the hidden layer neurons. [ 20 ] Table 1 Descriptive analysis of the input variables No. Variable code Variable description Median (Interquartile range) P-value Training group (1) Validation group (0) Test group (2) 1 Age bracket (year) Age size 68 (58–76) 69.5 (57–77) 72 (59–80) 0.425 2 Gender Gender (man, n%) 64.4% 53.7% 58.5% 0.178 5 ADL Activities of daily living 55 (35–80) 65 (30–80) 50 (20–85) 0.422 6 WBC (×10 9 /L) White blood cell 6.19 (5.225–7.43) 6.115 (4.98–7.8) 6.28 (5.49–7.39) 0.947 7 NEUT (×10 9 /L) Neutrophil count 3.665 (2.92–4.7) 3.635 (2.7–4.8) 3.665 (2.93–4.71) 0.873 8 LYM (×10 9 /L) Lymphocyte count 1.66 (1.37–2.08) 1.63 (1.4–2.13) 1.71 (1.25–2.2) 0.984 9 hs-CRP (mg/L) Hypersensitive C-reactive protein 1.96 (0.905–5.35) 2.095 (0.89–5.59) 2.095 (0.84–5.35) 0.993 10 Alb (g/L) Albumin 41 (38.1–43) 40 (37.3–42) 40 (37–43) 0.144 11 TG (mmol/L) Triglyceride 1.26 (0.95–1.725) 1.17 (0.93–1.56) 1.39 (1.02–1.82) 0.098 12 Hb (g/L) Hemoglobin 125 (115–135) 121.5 (111–135) 125 (117–134) 0.583 13 Stroke quality Stroke quality 75.3% 62.2% 75.6% 0.183 14 Stroke position Stroke position 22.9% 14.6% 31.7% 0.059 15 COPD Chronic obstructive pulmonary disease 2.4% 1.2% 3.7% 0.595 16 DM Diabetes mellitus 31.8% 43.9% 35. 4% 0.127 17 Hypertension Hypertension 73.6% 79.3% 74.4% 0.581 18 AF Atrial fibrillation 8.9% 9.8% 7.3% 0.851 19 Smoking history Smoking history 7.9% 6.1% 4.9% 0.603 20 Drinking history Drinking history 5.5% 3.7% 2.4% 0.460 21 DOC Disorders of consciousness 6.8% 8.5% 9.8% 0.648 22 Tracheotomy Tracheotomy 4.8% 2.4% 9.8% 0.093 23 Nasal feeding tube Nasal feeding tube 14.4% 14.6% 20.7% 0.363 24 Dysphagia Dysphagia 17.5% 11.0% 20.7% 0.226 The test population consisted of 85 individuals diagnosed with SAP and 371 individuals without SAP. The output of the hidden layer neurons, which corresponds to the neurons input into the output layer, can be expressed as \(\:\text{s}\left(\text{l}\right)\) = \(\:{\sum\:}_{j=1}^{R1}V\left({\iota\:},j\right)\text{*}y\left(j\right)-b2\left({\iota\:}\right);\:o\left({\iota\:}\right)=f\left(s\right({\iota\:}\left)\right),\:\:{\iota\:}=\text{1,2},...,R3.\) The connection weight between the neurons in the hidden layer ( j ) and the neurons in the output layer V ( \(\:{\iota\:}\) , j ) is represented in the formula, and the threshold value for output layer neurons is denoted by b 2( \(\:{\iota\:}\) ). To improve the processing speed of the network while preserving the performance, a pruning technique was employed at the conclusion of the training process. Weights that fell below the threshold value (0.5) for both input and hidden units were eliminated, taking into account the number of hidden units. [ 20 ] The output layer contained a single neuron that signified the presence of SAP, with a value of 1 denoting a positive response and a value of 0 representing a negative response. Based on the 7:2:2 ratio, the training, validation, and testing rates were approximately 64%, 18%, and 18%, respectively. Logistic regression model For comparison with the ANN model, a traditional LR model was also developed to forecast the probability of SAP. The LR model was used to calculate the Pi (probability of response) based on independent variables X 1, X 2, ..., and Xn : $$\:\:\text{l}\text{o}\text{g}\text{i}\text{t}\left(Pi\right)=log\left(\frac{Pi\:}{1-Pi\:}\right)=\beta\:0\hspace{0.17em}+\hspace{0.17em}\beta\:1X1\:+···+\:\beta\:nXn,$$ where β0 represents the y-intercept, and β i denotes the coefficient of the corresponding independent variable. [ 21 ] We randomly divided the dataset into two parts, the training set and the test set, with a ratio of 1:1. The LR model started with the same set of input variables as the ANN model. The outcome was defined as a dichotomous variable with a value of 1 (representing the presence of SAP or a positive outcome) or 0 (denoting the absence of SAP or a negative outcome). Statistical analysis Data were organized and statistically analyzed using SPSS version 26.0 (IBM Corp, Armonk, NY, USA). Model performance was analyzed based on the accuracy, positive predictive value (PPV), negative predictive value (NPV), sensitivity (SEN), specificity (SPE), and area under the curve (AUC) of the receiver operating characteristic (ROC) curve. Statistically significant differences were indicated by P < 0.05. Continuous variables were denoted by M (Q25–Q75). Significant differences between groups were assessed using chi-square test. Matlab 2017a was used to evaluate the abilities of the LR and ANN models to predict the incidence of SAP based on the ROC curves. Results ANN model prediction of SAP risk The model for ANN was built using the training dataset. The characteristics of the training, validation, and test datasets are summarized in Table 1 . None of the 24 variables showed significant variations across the three datasets ( P > 0.05), indicating a well-balanced distribution of clinical characteristics. Upon analysis of the training BP ANN model, Alb, ADL, Hb, and hs-CRP emerged as the most influential factors among the 24 independent variables for SAP, with respective normalized importance values of 100%, 75.8%, 57.4%, and 46.6% (Fig. 2 ). The test set was modeled using a BP neural network with a SEN of 83.53% and a SPE of 85.18% (Table 2 ). During our analysis, we used the cumulative gains graph to compare the forecasting precision of SAP as well as non-SAP models. The BP ANN model had a higher degree of fitting for SAP than for non-SAP, indicating that it is better suited to predicting SAP than non-SAP (Fig. 3 ). Table 2 Comparison of the ANN and LR models for predicting SAP following stroke in the test set Variable ANN Model LR Model Difference between models (95% CI) P -value Sensitivity 83.53% 71.76% 11.77% (− 2.85 − 26.39%) < 0.001 Specificity 85.18% 84.91% 0.27% (− 10.66 − 11.20%) 1 PPV 56.29% 52..08% 4.21% (− 17.09 − 25.51%) 0.232 NPV 95.77% 92.94% 2.83% (− 1.73 − 7.39%) 0.042 AUC 0.920 (0.891 − 0.943) 0.853 (0.817 − 0.884) 0.0244 (0.0187 − 0.114) 0.0064 Abbreviation: ANN, artificial neural network; PPV, positive predictive value; NPV, negative predictive value; AUC, area under the receiver operating characteristic curve. Logistic regression analyses One-way LR analysis indicated that a total of 24 factors were significantly associated with SAP, as shown in Table 3 . Among these factors, age, ADL, NEUT, LYM, hs-CRP, Alb, TG, Hb, stroke quality, COPD, AF, DOC, tracheotomy, and dysphagia showed statistically significant differences between SAP patients and non-SAP patients ( P < 0.05). Multivariate LR analysis indicated that age, ADL, NEUT, LYM, hs-CRP, Alb, TG, Hb, stroke quality, COPD, AF, DOC, tracheotomy, and dysphagia were significantly associated with SAP ( P < 0.05; Table 3 ). Applying the LR model to the test set with a classification threshold of 0.5 yielded a SEN of 71.76% (Table 2 ) and a SPE of 84.91% . Table 3 Univariate analysis of factors affecting SAP following stroke Variable Univariate Analysis Multivariate Analysis SAP Medians (IQR) Non-SAP Medians (IQR) x 2 -Test (df) H-value (df) P -value B S.E. Sig. Exp( B ) 95% CI for Exp(B) Age (year) 73 (59–82) 68 (58–75) –3.05 0.002 −0.011 0.013 0.408 0.989 0.964–1.015 Gender Male 20% 80% 0.884 0.347 - Female 16.50% 83.50% ADL 20 (10–30) 65 (45–85) −11.035 < 0.0001 −0.054 0.009 0 0.947 0.931–0.964 WBC (×10 9 /L) 6.46 (5.22–8.19) 6.16 (5.19–7.34) −1.511 0.131 - NEUT (×10 9 /L) 3.87 (3.2–5.43) 3.62 (2.85–4.63) −-2.381 0.017 0.067 0.078 0.39 1.069 0.918–1.245 LYM (×10 9 /L) 1.54 (1.22–1.81) 1.72 (1.39–2.155) −3.003 0.003 −0.143 0.269 0.594 0.867 0.512–1.467 hs-CRP (mg/L) 5.94 (1.67–16.33) 1.75 (0.84–4.365) −5.046 < 0.0001 −0.006 0.011 0.551 0.994 0.973–1.015 Alb (g/L) 37.3 (34–40) 41 (39–43.45) −7.699 < 0.0001 −0.137 0.05 0.006 0.872 0.791–0.962 TG (mmol/L) 1.05 (0.86–1.51) 1.31 (1.005–1.78) −2.988 0.003 −0.193 0.27 0.475 0.824 0.485–1.400 Hb (g/L) 117 (108–127) 127 (116–136) −4.585 < 0.0001 −0.005 0.011 0.672 0.995 0.974–1.017 Stroke quality CH = 1 28.40% 71.60% 6.726 0.035 −0.491 0.462 0.287 0.612 0.248–1.512 CI = 2 15.90% 84.10% Mixed = 3 22.40% 77.60% Stroke position unilateral = 1 16.60% 83.40% 2.477 0.294 - bilateral = 2 22.90% 77.10% brainstem or cerebellum = 3 21.80% 78.20% COPD Yes 54.50% 45.50% 9.582 0.002 2.744 1.078 0.011 15.551 1.879–128.737 No 17.80% 82.20% DM Yes 13.90% 86.10% 3.546 0.06 - No 21.10% 78.90% Hypertension Yes 18.80% 81.20% 0.015 0.904 - No 18.30% 81.70% AF Yes 52.50% 47.50% 33.146 < 0.0001 1.32 0.529 0.013 3.744 1.327–10.564 No 15.40% 84.60% Smoking history Yes 18.80% 81.30% 0 0.987 - No 18.60% 81.40% drinking history Yes 19.00% 81.00% 0.002 0.961 - No 18.60% 81.40% DOC Yes 65.70% 34.30% 55.392 < 0.0001 0.118 0.598 0.843 1.125 0.349–3.632 No 14.70% 85.3 Tracheotomy Yes 95.80% 4.20% 99.537 < 0.0001 2.763 1.111 0.013 15.855 1.796–139.997 No 14.40% 85.60% Nasal feeding tube Yes 80.30% 19.70% 210.69 < 0.0001 - No 7.30% 92.70% Dysphagia Yes 35.10% 64.90% 16.479 < 0.0001 0.36 0.386 0.351 1.434 0.673–3.057 No 15.30% 84.70% ADL: activities of daily living; WBC: white cell count; NEUT: neutrophil count; LYM: lymphocyte count; hs-CRP: hypersensitive C-reactive protein; Alb: albumin; TG: triglyceride; Hb: hemoglobin; CH: cerebral hemorrhage CI: cerebral infarction; COPD: chronic obstructive pulmonary disease; DM: diabetes mellitus; AF: atrial fibrillation; DOC: disorders of consciousness Comparison of the LR and ANN models The prediction results of both the LR and ANN models are summarized in Table 2 . A significant difference in SEN was observed between the models ( P < 0.001). The test dataset used to identify SAP was also employed to construct the ROC curves for both models. The AUC values of the LR and ANN models were 0.853 (0.817–0.884) and 0.920 (0.891–0.943), respectively, indicating that the ANN model was more accurate overall (Fig. 4 ; P = 0.0064). Discussion SAP is an important determinant of adverse clinical outcomes and mortality after stroke. [ 22 ] The mortality rate of stroke patients with pneumonia is 5.58 times that of stroke patients without pneumonia. [ 23 ] Infection is a risk factor for stroke and plays an important role in the prognosis after stroke. [ 24 ] Infection, including SAP, is a risk factor for stroke and plays a crucial role in post-stroke prognosis. Due to its higher prevalence and challenging treatment, SAP has a more significant influence on clinical outcomes. [ 25 ] At present, the assessment of SAP risk remains challenging in clinical practice. Therefore, it is necessary to identify patients at risk for SAP at an early stage, which may facilitate the implementation of monitoring and preventive measures in routine clinical care. In this study, we compared ANN and LR models for identifying stroke patients at risk of SAP. ANNs, which do not suffer from the uncertainties inherent in linearized models, can be used to analyze complex nonlinear relationships in the medical field with the goal of improving system performance. [ 26 ] Consequently, ANN models are favored over LR models because of their superior predictive accuracy when addressing complex nonlinear problems. [ 27 , 28 ] ANNs have shown promise in numerous applications including biochemical analysis, disease prediction, and diagnostic systems. [ 29 , 30 ] In the current study, the ANN model exhibited greater accuracy in predicting the occurrence of SAP compared to the LR model. The ANN model for predicting SAP outperformed the LR approach in terms of SEN, SPE, PPV, and NPV. We also identified several predictors of SAP, including dysphagia and C-reactive protein, in alignment with previous studies suggesting that these factors are associated with the incidence of SAP. [ 31 , 32 ] We incorporated additional variables into the ANN model to enhance its predictive capacity. Notably, we found a significant association between Alb and SAP. Albumin, which has been shown to have a high diagnostic value, is significantly linked to the development of SAP in older patients; the risk of SAP increases as the albumin level declines. [ 33 – 35 ] Albumin has an antioxidant effect; thus, a reduction in Alb may lead to increased oxidative stress, increased cell damage and inflammation, and a weakening of the blood–brain barrier, making it easier for infectious agents to enter and increasing the risk of secondary infection after stroke. [ 36 ] Additionally, we identified a lower ADL score as an independent risk factor for pneumonia. A low ADL score typically indicates a high level of nursing dependence; among stroke patients, those with relatively low ADL scores upon hospital admission were found to have a higher likelihood of developing SAP. [ 37 ] In the present study, we also showed that low hemoglobin levels were an independent risk factor for SAP. Low hemoglobin concentrations are associated with a reduced capacity for oxygen transport, leading to an increased likelihood of infarction in regions with compromised blood flow.Decreased hemoglobin concentrations are associated with a diminished capacity for oxygen transport, resulting in increased likelihood of infarction in regions with compromised blood flow. [ 38 ] Stroke patients may be nutritionally imbalanced, further increasing the rate of infection. [ 39 ] CRP has the potential to predict the occurrence of SAP [ 40 ] because elevated CRP levels are closely linked to the progression of SAP. [ 41 ] The levels of acute-phase plasma proteins such as CRP increase significantly during the acute phase of an inflammatory reaction. [ 25 ] When CRP levels are elevated, it reflects the activation of the inflammatory response in the body, which may lead to the growth of pathogens such as bacteria and viruses in the lungs, which can lead to pneumonia. In stroke patients, the inflammatory response is exacerbated by factors such as weakened immunity and impaired coughing and swallowing reflexes, which further increases the risk of SAP [ 42 ] . It is important to recognize the constraints inherent in the present study. First, this was a single-center retrospective study, which inherently restricts the generalizability of the findings. The presence of missing values in some analyzed variables also limits the utility of the findings. Second, our findings may be subject to potential bias due to insufficient available data and lack of external validation. Consequently, additional multicenter, prospective studies are needed to overcome these limitations. Conclusions We successfully developed and validated an ANN model to identify stroke patients at risk for SAP during their preliminary evaluation or at the point of entry into the emergency department. The application of this predictive model promises significant improvements in the early identification of SAP cases, enabling timely interventions that could mitigate the progression of SAP. Future efforts will focus on further optimizing the model to enhance its clinical applicability in diverse healthcare settings. Declarations Funding This work was supported by the Natural Key Projects of Research Projects in Anhui Province Universities (2023AH050784). Author contributions Ting Wang, Linli Yuan, Min You, Juan Yuan and Sijing Peng conceived of and designed the study;Ting Wang, Chunbiao Li conducted the experiments and authored the manuscript; Ting Wang, Chunbiao Liworked on the data integration,preprocessing and analysis; Yi Liu, Min Yang, Yaling Fan, Qinsi Tong and Dajin Li collected the data;Min You, Juan Yuan oversaw the project, coordinated;All authors read and approved the final manuscript. Conflicts of interest There are no conflicts of interest to be declared. Ethics approval and consent to participate This study was approved by the Ethics Committee of the Second Affiliated Hospital of Anhui University of Traditional Chinese Medicine (2021-zj-24). In addition, informed consent was omitted for all participants as the study was retrospective in nature, and the research process was in accordance with the Declaration of Helsinki. Acknowledgments The authors thank AiMi Academic Services (www.aimieditor.com) for English language editing and review services. Declaration of Generative AI and AI-assisted technologies in the writing process References Smith CJ, Kishore AK, Vail A, et al. Diagnosis of Stroke-Associated Pneumonia: Recommendations from the Pneumonia in Stroke Consensus Group. Stroke. 2015;46:2335–40. Westendorp WF, Nederkoorn PJ, Vermeij J-D, et al. Post-stroke infection: a systematic review and meta-analysis. BMC Neurol. 2011;11:110. Badve MS, Zhou Z, van de Beek D, et al. Frequency of post-stroke pneumonia: Systematic review and meta-analysis of observational studies. Int J Stroke Off J Int Stroke Soc. 2019;14:125–36. Teh WH, Smith CJ, Barlas RS, et al. Impact of stroke-associated pneumonia on mortality, length of hospitalization, and functional outcome. Acta Neurol Scand. 2018;138:293–300. Ali AN, Howe J, Majid A, et al. The economic cost of stroke-associated pneumonia in a UK setting. Top Stroke Rehabil. 2018;25:214–23. Kwon H-M, Jeong S-W, Lee S-H, et al. The pneumonia score: A simple grading scale for prediction of pneumonia after acute stroke. Am J Infect Control. 2006;34:64–8. Hoffmann S, Malzahn U, Harms H, et al. Development of a Clinical Score (A2DS2) to Predict Pneumonia in Acute Ischemic Stroke. Stroke. 2012;43:2617–23. Smith CJ, Bray BD, Hoffman A, et al. Can a Novel Clinical Risk Score Improve Pneumonia Prediction in Acute Stroke Care? A UK Multicenter Cohort Study. J Am Heart Assoc. 2015;4:e001307. Kumar S, Marchina S, Massaro J, et al. ACDD4 score: A simple tool for assessing risk of pneumonia after stroke. J Neurol Sci. 2017;372:399–402. Huang G-Q, Lin Y-T, Wu Y-M et al. Individualized Prediction Of Stroke-Associated Pneumonia For Patients With Acute Ischemic Stroke. CIA. 2019; Volume 14:1951–1962. Jiang F, Dong L, Dai Q. Electrical resistivity imaging inversion: An ISFLA trained kernel principal component wavelet neural network approach. Neural Netw Off J Int Neural Netw Soc. 2018;104:114–23. Fei Y, Gao K, Li W-Q. Artificial neural network algorithm model as powerful tool to predict acute lung injury following to severe acute pancreatitis. Pancreatol Off J Int Assoc Pancreatol IAP Al. 2018;18:892–9. Chung C-C, Chan L, Bamodu OA, et al. Artificial neural network based prediction of postthrombolysis intracerebral hemorrhage and death. Sci Rep. 2020;10:20501. Hou J, Fu S, Wang X, et al. A noninvasive artificial neural network model to predict IgA nephropathy risk in Chinese population. Sci Rep. 2022;12:8296. Tafeit E, Reibnegger G. Artificial neural networks in laboratory medicine and medical outcome prediction. Clin Chem Lab Med. 1999;37:845–53. Naidu SMM, Pandey PC, Bagal UR, et al. Beat-to-beat estimation of stroke volume using impedance cardiography and artificial neural network. Med Biol Eng Comput. 2018;56:1077–89. Moon S, Ahmadnezhad P, Song H-J, et al. Artificial neural networks in neurorehabilitation: A scoping review. NeuroRehabilitation. 2020;46:259–69. Lu F, Liang Y, Wang X, et al. Prediction of amorphous forming ability based on artificial neural network and convolutional neural network. Comput Mater Sci. 2022;210:111464. Fei Y, Gao K, Li W. Prediction and evaluation of the severity of acute respiratory distress syndrome following severe acute pancreatitis using an artificial neural network algorithm model. HPB. 2019;21:891–7. Fei Y, Gao K, Li W. Artificial neural network algorithm model as powerful tool to predict acute lung injury following to severe acute pancreatitis. Pancreatology. 2018;18:892–9. Shariatnia S, Ziaratban M, Rajabi A, et al. Modeling the diagnosis of coronary artery disease by discriminant analysis and logistic regression: a cross-sectional study. BMC Med Inf Decis Mak. 2022;22:85. Patel UK, Kodumuri N, Dave M, et al. Stroke-Associated Pneumonia: A Retrospective Study of Risk Factors and Outcomes. Neurologist. 2020;25:39–48. Westendorp WF, Nederkoorn PJ, Vermeij J-D, et al. Post-stroke infection: A systematic review and meta-analysis. BMC Neurol. 2011;11:110. Elkind MSV, Boehme AK, Smith CJ, et al. Infection as a Stroke Risk Factor and Determinant of Outcome After Stroke. Stroke. 2020;51:3156–68. Hilker R, Poetter C, Findeisen N, et al. Nosocomial Pneumonia After Acute Stroke: Implications for Neurological Intensive Care Medicine. Stroke. 2003;34:975–81. Hu J, Liu L, Wang Y, et al. Precision motion control of a small launching platform with disturbance compensation using neural networks: INTELLIGENT CONTROL IN LAUNCHING TECHNOLOGY. Int J Adapt Control Signal Process. 2017;31:971–84. Ottenbacher KJ, Linn RT, Smith PM, et al. Comparison of logistic regression and neural network analysis applied to predicting living setting after hip fracture. Ann Epidemiol. 2004;14:551–9. Dreiseitl S, Ohno-Machado L. Logistic regression and artificial neural network classification models: a methodology review. J Biomed Inf. 2002;35:352–9. Chun-An Cheng null. Hung-Wen Chiu null. An artificial neural network model for the evaluation of carotid artery stenting prognosis using a national-wide database. Annu Int Conf IEEE Eng Med Biol Soc IEEE Eng Med Biol Soc Annu Int Conf 2017; 2017:2566–2569. Rau H-H, Hsu C-Y, Lin Y-A, et al. Development of a web-based liver cancer prediction model for type II diabetes patients by using an artificial neural network. Comput Methods Programs Biomed. 2016;125:58–65. Huang G-Q, Lin Y-T, Wu Y-M, et al. Individualized Prediction Of Stroke-Associated Pneumonia For Patients With Acute Ischemic Stroke. Clin Interv Aging. 2019;14:1951–62. Li X, Wu M, Sun C, et al. Using machine learning to predict stroke-associated pneumonia in Chinese acute ischaemic stroke patients. Eur J Neurol. 2020;27:1656–63. Wiedermann CJ. Hypoalbuminemia as Surrogate and Culprit of Infections. Int J Mol Sci. 2021;22:4496. Lv X-N, Shen Y-Q, Li Z-Q, et al. Neutrophil percentage to albumin ratio is associated with stroke-associated pneumonia and poor outcome in patients with spontaneous intracerebral hemorrhage. Front Immunol. 2023;14:1173718. Zawiah M, Khan AH, Abu Farha R, et al. Predictors of stroke-associated pneumonia and the predictive value of neutrophil percentage-to-albumin ratio. Postgrad Med. 2023;135:681–9. Halliwell B. Albumin—An important extracellular antioxidant? Biochem Pharmacol. 1988;37:569–71. Watanabe S, Shimozato K, Oh-Shige H, et al. Examination of factors associated with aspiration pneumonia following stroke. Oral Sci Int. 2014;11:15–21. Naess H, Logallo N, Waje-Andreassen U, et al. U‐shaped relationship between hemoglobin level and severity of ischemic stroke. Acta Neurol Scand. 2019;140:56–61. Song X, He Y, Bai J, et al. A nomogram based on nutritional status and A2DS2 score for predicting stroke-associated pneumonia in acute ischemic stroke patients with type 2 diabetes mellitus: A retrospective study. Front Nutr. 2022;9:1009041. Li Y, Zhao L, Liu Y, et al. Novel Predictors of Stroke-Associated Pneumonia: A Single Center Analysis. Front Neurol. 2022;13:857420. Kalra L, Smith CJ, Hodsoll J, et al. Elevated C-reactive protein increases diagnostic accuracy of algorithm-defined stroke-associated pneumonia in afebrile patients. Int J Stroke. 2019;14:167–73. Kalra L, Smith CJ, Hodsoll J, et al. Elevated C-reactive protein increases diagnostic accuracy of algorithm-defined stroke-associated pneumonia in afebrile patients. Int J Stroke. 2019;14:167–73. Additional Declarations No competing interests reported. Supplementary Files highlights.docx Cite Share Download PDF Status: Posted Version 1 posted 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-4754561","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":338704475,"identity":"4fce60c4-5354-4546-a6ff-f10b86aa2140","order_by":0,"name":"Ting Wang","email":"","orcid":"","institution":"The Second Affiliated Hospital of Anhui University of Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Ting","middleName":"","lastName":"Wang","suffix":""},{"id":338704476,"identity":"3abcd8a8-8fd2-4ed2-9200-6a277594ba6c","order_by":1,"name":"Chunbiao Li","email":"","orcid":"","institution":"Anhui University of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Chunbiao","middleName":"","lastName":"Li","suffix":""},{"id":338704477,"identity":"05de338a-d0df-487c-bb22-4f3d828e1c92","order_by":2,"name":"Linli Yuan","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA90lEQVRIiWNgGAWjYLCCBwYMMmzMzIcffGwAcRkbDxDUkmDAwMPGzpZmOLOBQQKopYEILQwMPAz8PArSvGAtDAx4tRjcSH72IKHgDg8fMw+Dse0Omzrd9sNAW2psonFrSTM3SDB4xsPGzHvgce6ZNAmzM4lALcfSchtwakkwk0gwOAzUwpdgnNt2WMLsAFALY8NhPFrSv0G18BhIW4K0nH9ISEuOGUILI0jLDQK2SJ55UwbVAgzk3rY0yW03gLYk4PEL3/H0bRIf/hyWk+8/fPjBzzYbfrPz6Q8ffKixwalF4QBW4QQcykFAHpdZo2AUjIJRMArgAABvt15c9fn4zAAAAABJRU5ErkJggg==","orcid":"","institution":"The Second Affiliated Hospital of Anhui University of Chinese Medicine","correspondingAuthor":true,"prefix":"","firstName":"Linli","middleName":"","lastName":"Yuan","suffix":""},{"id":338704478,"identity":"bf6523cd-6a16-493f-be92-7a46aebed581","order_by":3,"name":"Min You","email":"","orcid":"","institution":"The Second Affiliated Hospital of Anhui University of Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Min","middleName":"","lastName":"You","suffix":""},{"id":338704479,"identity":"ec9d6c90-45bf-4cd0-a433-556a0ea4d7b8","order_by":4,"name":"Juan Yuan","email":"","orcid":"","institution":"Anhui University of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Juan","middleName":"","lastName":"Yuan","suffix":""},{"id":338704480,"identity":"51302d94-fb73-413b-ac4f-e4b104417282","order_by":5,"name":"Sijing Peng","email":"","orcid":"","institution":"Anhui University of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Sijing","middleName":"","lastName":"Peng","suffix":""},{"id":338704481,"identity":"bff4a97e-e228-46e5-bad9-907178e51299","order_by":6,"name":"Yi Liu","email":"","orcid":"","institution":"Anhui University of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Yi","middleName":"","lastName":"Liu","suffix":""},{"id":338704482,"identity":"178f8d3a-e78f-4c66-9cd4-784badeba7db","order_by":7,"name":"Min Yang","email":"","orcid":"","institution":"The Second Affiliated Hospital of Anhui University of Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Min","middleName":"","lastName":"Yang","suffix":""},{"id":338704483,"identity":"a8924070-71fb-4ec1-9807-d60eb4ec4766","order_by":8,"name":"Yaling Fan","email":"","orcid":"","institution":"Anhui University of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Yaling","middleName":"","lastName":"Fan","suffix":""},{"id":338704484,"identity":"5ff67b98-3443-4cbd-b3ef-cd7ce463032e","order_by":9,"name":"Qinsi Tong","email":"","orcid":"","institution":"The Second Affiliated Hospital of Anhui University of Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Qinsi","middleName":"","lastName":"Tong","suffix":""},{"id":338704485,"identity":"89ffbb20-8c7c-4535-965d-673477e974bf","order_by":10,"name":"Dajin Li","email":"","orcid":"","institution":"Anhui University of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Dajin","middleName":"","lastName":"Li","suffix":""}],"badges":[],"createdAt":"2024-07-17 08:01:33","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4754561/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4754561/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":62732656,"identity":"fcd5355a-77fb-4f23-8c19-ceaafbd13099","added_by":"auto","created_at":"2024-08-18 23:46:30","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":122508,"visible":true,"origin":"","legend":"\u003cp\u003eSchematic topology of the ANN model.\u003c/p\u003e","description":"","filename":"Fig.1.png","url":"https://assets-eu.researchsquare.com/files/rs-4754561/v1/f2f8395e2a2aea156d050714.png"},{"id":62733183,"identity":"c6d53fc7-39da-40c8-b78d-6b7425548deb","added_by":"auto","created_at":"2024-08-18 23:54:30","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":146403,"visible":true,"origin":"","legend":"\u003cp\u003eThe value of artificial neural networks modeling in predicting indicators of SAP.\u003c/p\u003e","description":"","filename":"Fig.2.png","url":"https://assets-eu.researchsquare.com/files/rs-4754561/v1/33e498b792d26e6b2e683930.png"},{"id":62732660,"identity":"8a436806-e8be-486d-90f5-51fd78c896de","added_by":"auto","created_at":"2024-08-18 23:46:30","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":199924,"visible":true,"origin":"","legend":"\u003cp\u003eCumulative gains plot of the ANN model for predicting SAP in stroke patients (0 means non-SAP, 1 means SAP).\u003c/p\u003e","description":"","filename":"Fig.3.png","url":"https://assets-eu.researchsquare.com/files/rs-4754561/v1/a4f822f3bcd50cf7c7e2802a.png"},{"id":62732658,"identity":"0882c0f2-3fe7-42ab-a3f0-bbe6060ce2df","added_by":"auto","created_at":"2024-08-18 23:46:30","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":144001,"visible":true,"origin":"","legend":"\u003cp\u003eReceiver operating characteristic (ROC) curve for predicting SAP following stroke. LR = logistic regression; ANN = artificial neural network.\u003c/p\u003e","description":"","filename":"Fig.4.png","url":"https://assets-eu.researchsquare.com/files/rs-4754561/v1/138fe180f7af2cf929f96eda.png"},{"id":74431192,"identity":"a967f12b-8429-49af-b751-b599ed0c7c21","added_by":"auto","created_at":"2025-01-22 08:47:18","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1745840,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4754561/v1/3ec193fe-cbe7-40ce-a05b-2f8ec1d60a62.pdf"},{"id":62733184,"identity":"c7cd7372-4460-4371-8ff1-b4659b76864f","added_by":"auto","created_at":"2024-08-18 23:54:30","extension":"docx","order_by":7,"title":"","display":"","copyAsset":false,"role":"supplement","size":11543,"visible":true,"origin":"","legend":"","description":"","filename":"highlights.docx","url":"https://assets-eu.researchsquare.com/files/rs-4754561/v1/9ddcc6463267fc81fd72f094.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Prediction and evaluation of the risk of Stroke-associated pneumonia using an artificial neural network model","fulltext":[{"header":"Introduction","content":"\u003cp\u003eStroke-associated pneumonia (SAP), which occurs within seven days of onset in non-ventilated stroke patients, is one of the most common complications of cerebrovascular accidents.\u003csup\u003e[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]\u003c/sup\u003e Surveys indicate that between 5% and 65% of stroke patients develop SAP,\u003csup\u003e[\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]\u003c/sup\u003e which can increase the mortality rate, prolong the hospital stay, and impose a significant economic burden on the patient.\u003csup\u003e[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]\u003c/sup\u003e As a result, the early identification and assessment of SAP risk is critical for improving patient prognosis and reducing healthcare costs.\u003c/p\u003e \u003cp\u003eNumerous risk prediction models with varying degrees of predictive effectiveness have been developed for predicting the occurrence of SAP.\u003csup\u003e[\u003cspan additionalcitationids=\"CR7 CR8 CR9\" citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]\u003c/sup\u003e However, these models fail to consider the comprehensive range of risk factors and associated variables. Furthermore, the accuracy and stability of typical statistical methods such as logistic regression (LR) are compromised when dealing with nonlinear relationships and high-dimensional interactions.\u003c/p\u003e \u003cp\u003eArtificial neural networks (ANNs), also termed multi-layer perceptrons, are deep neural networks that can be used to model problems in which the relationships between causal factors and responses are complex.\u003csup\u003e[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]\u003c/sup\u003e In recent years, ANNs have been increasingly used as powerful machine learning tools in risk prediction models.\u003csup\u003e[\u003cspan additionalcitationids=\"CR13\" citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]\u003c/sup\u003e ANNs are capable of learning and simulating complex nonlinear relationships between variables; more importantly, they have the ability to self-learn, which allows them to optimize the model by progressively adjusting the weights within the network. Consequently, the resulting ANNs may demonstrate superior performance compared with traditional statistical models when identifying risk factors for SAP.\u003csup\u003e[\u003cspan additionalcitationids=\"CR16\" citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eIn this study, we developed an ANN model to predict the risk of SAP in stroke patients and compared its predictive performance with that of a traditional LR model. Our goal was to construct an accurate prediction tool that can effectively identify patients at high risk of SAP, thereby providing support for clinical decision-making.\u003c/p\u003e "},{"header":"Methods","content":"\u003ch3\u003eGeneral information\u003c/h3\u003e\n\u003cp\u003eAll stroke patients were enrolled in the clinical departments of our hospital from July 2017 to April 2022. The inclusion criteria were as follows: (1) age\u0026thinsp;\u0026ge;\u0026thinsp;18; and (2) stroke diagnosis was confirmed via computed tomography or magnetic resonance imaging of the skull. The exclusion criteria were as follows: (1) patients with pre-existing pneumonia prior to admission; and (2) a history of hematological diseases, malignancy, immunosuppressant treatment, or severe hepatic or renal dysfunction. SAP was diagnosed based on clinical, laboratory, and radiological data by the treating physician and recorded on the medical chart. International guidelines were followed by all the patients in terms of standard medical treatment.\u003c/p\u003e \u003cp\u003eA total of 456 patients (280 men and 176 women) participated in this study. The protocol for the study was formulated in accordance with the guidelines of the institutional ethical committee. Informed consent was given by all participants, and the research was carried out in accordance with the ethical principles outlined in the Helsinki Declaration.\u003c/p\u003e \u003cp\u003eAfter conducting a thorough examination of literature, we selected 24 factors to incorporate into the ANN model: demographic factors (gender and age); medical history (chronic obstructive pulmonary disease, high blood pressure, diabetes, atrial fibrillation, smoking history, and drinking history); laboratory data (white blood cell count, neutrophil count, lymphocyte count, hypersensitive C-reactive protein [CRP] level, albumin level, triglyceride level, and hemoglobin level); and hospitalization characteristics (activities of daily living score, stroke quality, stroke position, disorders of consciousness, and dysphagia).\u003c/p\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eANN model\u003c/h2\u003e \u003cp\u003eThe ANN model consisted of three parts: an input layer, a hidden layer, and an output layer(Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003csup\u003e[\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]\u003c/sup\u003e The input layer was a feed-forward multilayer structure consisting of nonlinear neurons (i.e., perceptrons), and every layer consisted of anthropomorphic neurons modeled by nodes.\u003csup\u003e[\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]\u003c/sup\u003e The ANN model used an integrated set of 24 input variables; Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e gives an overview of these input variables along with their means, standard deviations, and frequencies. If the inputs to the input hierarchy are set to x(1), x(2),..., x(R), the output of the input layer corresponding to the input of the hidden layer neurons can be represented as \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:s\\left(i\\right)={\\sum\\:}_{j=1}^{R1}W\\left(i,j\\right)\\text{*}x\\left(j\\right)-b1\\left(i\\right);\\:y\\left(i\\right)=\\text{f}\\left(s\\right(i\\left)\\right),i=\\text{1,2},...,R2,\\)\u003c/span\u003e\u003c/span\u003e where \u003cem\u003eW\u003c/em\u003e(\u003cem\u003ei\u003c/em\u003e, \u003cem\u003ej\u003c/em\u003e) denotes the connection weight of the input layer neurons (\u003cem\u003ei\u003c/em\u003e) and the hidden layer neurons (\u003cem\u003ej\u003c/em\u003e), and \u003cem\u003eb\u003c/em\u003e1(\u003cem\u003ej\u003c/em\u003e) represents the threshold of the hidden layer neurons.\u003csup\u003e[\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]\u003c/sup\u003e\u003c/p\u003e \u003cp\u003e \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\u003eDescriptive analysis of the input variables\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eNo.\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eVariable code\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eVariable description\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c6\" namest=\"c4\"\u003e \u003cp\u003eMedian (Interquartile range)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e\u003cem\u003eP-value\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eTraining group (1)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eValidation group (0)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eTest group (2)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAge bracket (year)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAge size\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e68 (58\u0026ndash;76)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e69.5 (57\u0026ndash;77)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e72 (59\u0026ndash;80)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.425\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGender\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eGender (man, n%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e64.4%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e53.7%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e58.5%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.178\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eADL\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eActivities of daily living\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e55 (35\u0026ndash;80)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e65 (30\u0026ndash;80)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e50 (20\u0026ndash;85)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.422\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eWBC (\u0026times;10\u003csup\u003e9\u003c/sup\u003e/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eWhite blood cell\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e6.19 (5.225\u0026ndash;7.43)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e6.115 (4.98\u0026ndash;7.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e6.28 (5.49\u0026ndash;7.39)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.947\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNEUT (\u0026times;10\u003csup\u003e9\u003c/sup\u003e/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNeutrophil count\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3.665 (2.92\u0026ndash;4.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3.635 (2.7\u0026ndash;4.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e3.665 (2.93\u0026ndash;4.71)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.873\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLYM (\u0026times;10\u003csup\u003e9\u003c/sup\u003e/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eLymphocyte count\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.66 (1.37\u0026ndash;2.08)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.63 (1.4\u0026ndash;2.13)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1.71 (1.25\u0026ndash;2.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.984\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ehs-CRP (mg/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eHypersensitive C-reactive protein\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.96 (0.905\u0026ndash;5.35)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e2.095 (0.89\u0026ndash;5.59)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e2.095 (0.84\u0026ndash;5.35)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.993\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAlb (g/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAlbumin\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e41 (38.1\u0026ndash;43)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e40 (37.3\u0026ndash;42)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e40 (37\u0026ndash;43)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.144\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTG (mmol/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTriglyceride\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.26 (0.95\u0026ndash;1.725)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.17 (0.93\u0026ndash;1.56)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1.39 (1.02\u0026ndash;1.82)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.098\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eHb (g/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eHemoglobin\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e125 (115\u0026ndash;135)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e121.5 (111\u0026ndash;135)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e125 (117\u0026ndash;134)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.583\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eStroke quality\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eStroke quality\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e75.3%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e62.2%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e75.6%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.183\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eStroke position\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eStroke position\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e22.9%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e14.6%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e31.7%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.059\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCOPD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eChronic obstructive pulmonary disease\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2.4%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.2%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e3.7%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.595\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eDiabetes mellitus\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e31.8%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e43.9%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e35. 4%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.127\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eHypertension\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eHypertension\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e73.6%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e79.3%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e74.4%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.581\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e18\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAF\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAtrial fibrillation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e8.9%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e9.8%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e7.3%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.851\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSmoking history\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSmoking history\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e7.9%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e6.1%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e4.9%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.603\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDrinking history\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eDrinking history\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e5.5%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3.7%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e2.4%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.460\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDOC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eDisorders of consciousness\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e6.8%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e8.5%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e9.8%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.648\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e22\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTracheotomy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTracheotomy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4.8%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e2.4%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e9.8%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.093\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e23\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNasal feeding tube\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNasal feeding tube\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e14.4%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e14.6%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e20.7%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.363\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e24\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDysphagia\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eDysphagia\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e17.5%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e11.0%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e20.7%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.226\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eThe test population consisted of 85 individuals diagnosed with SAP and 371 individuals without SAP. The output of the hidden layer neurons, which corresponds to the neurons input into the output layer, can be expressed as \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\text{s}\\left(\\text{l}\\right)\\)\u003c/span\u003e\u003c/span\u003e=\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\sum\\:}_{j=1}^{R1}V\\left({\\iota\\:},j\\right)\\text{*}y\\left(j\\right)-b2\\left({\\iota\\:}\\right);\\:o\\left({\\iota\\:}\\right)=f\\left(s\\right({\\iota\\:}\\left)\\right),\\:\\:{\\iota\\:}=\\text{1,2},...,R3.\\)\u003c/span\u003e\u003c/span\u003e The connection weight between the neurons in the hidden layer (\u003cem\u003ej\u003c/em\u003e) and the neurons in the output layer \u003cem\u003eV\u003c/em\u003e(\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\iota\\:}\\)\u003c/span\u003e\u003c/span\u003e,\u003cem\u003ej\u003c/em\u003e) is represented in the formula, and the threshold value for output layer neurons is denoted by \u003cem\u003eb\u003c/em\u003e2(\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\iota\\:}\\)\u003c/span\u003e\u003c/span\u003e). To improve the processing speed of the network while preserving the performance, a pruning technique was employed at the conclusion of the training process. Weights that fell below the threshold value (0.5) for both input and hidden units were eliminated, taking into account the number of hidden units.\u003csup\u003e[\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]\u003c/sup\u003e The output layer contained a single neuron that signified the presence of SAP, with a value of 1 denoting a positive response and a value of 0 representing a negative response. Based on the 7:2:2 ratio, the training, validation, and testing rates were approximately 64%, 18%, and 18%, respectively.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003eLogistic regression model\u003c/h2\u003e \u003cp\u003eFor comparison with the ANN model, a traditional LR model was also developed to forecast the probability of SAP. The LR model was used to calculate the \u003cem\u003ePi\u003c/em\u003e (probability of response) based on independent variables \u003cem\u003eX\u003c/em\u003e1, \u003cem\u003eX\u003c/em\u003e2, ..., and \u003cem\u003eXn\u003c/em\u003e:\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\:\\:\\text{l}\\text{o}\\text{g}\\text{i}\\text{t}\\left(Pi\\right)=log\\left(\\frac{Pi\\:}{1-Pi\\:}\\right)=\\beta\\:0\\hspace{0.17em}+\\hspace{0.17em}\\beta\\:1X1\\:+\u0026middot;\u0026middot;\u0026middot;+\\:\\beta\\:nXn,$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003ewhere β0 represents the y-intercept, and β\u003cem\u003ei\u003c/em\u003e denotes the coefficient of the corresponding independent variable.\u003csup\u003e[\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]\u003c/sup\u003e We randomly divided the dataset into two parts, the training set and the test set, with a ratio of 1:1. The LR model started with the same set of input variables as the ANN model. The outcome was defined as a dichotomous variable with a value of 1 (representing the presence of SAP or a positive outcome) or 0 (denoting the absence of SAP or a negative outcome).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003eStatistical analysis\u003c/h2\u003e \u003cp\u003eData were organized and statistically analyzed using SPSS version 26.0 (IBM Corp, Armonk, NY, USA). Model performance was analyzed based on the accuracy, positive predictive value (PPV), negative predictive value (NPV), sensitivity (SEN), specificity (SPE), and area under the curve (AUC) of the receiver operating characteristic (ROC) curve. Statistically significant differences were indicated by \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05. Continuous variables were denoted by M (Q25\u0026ndash;Q75). Significant differences between groups were assessed using chi-square test. Matlab 2017a was used to evaluate the abilities of the LR and ANN models to predict the incidence of SAP based on the ROC curves.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eANN model prediction of SAP risk\u003c/h2\u003e \u003cp\u003eThe model for ANN was built using the training dataset. The characteristics of the training, validation, and test datasets are summarized in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. None of the 24 variables showed significant variations across the three datasets (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026gt;\u0026thinsp;0.05), indicating a well-balanced distribution of clinical characteristics. Upon analysis of the training BP ANN model, Alb, ADL, Hb, and hs-CRP emerged as the most influential factors among the 24 independent variables for SAP, with respective normalized importance values of 100%, 75.8%, 57.4%, and 46.6% (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe test set was modeled using a BP neural network with a SEN of 83.53% and a SPE of 85.18% (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). During our analysis, we used the cumulative gains graph to compare the forecasting precision of SAP as well as non-SAP models. The BP ANN model had a higher degree of fitting for SAP than for non-SAP, indicating that it is better suited to predicting SAP than non-SAP (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e).\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 the ANN and LR models for predicting SAP following stroke in the test set\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" 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=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eANN Model\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eLR Model\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eDifference between models (95% CI)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cem\u003eP\u003c/em\u003e-value\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSensitivity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e83.53%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e71.76%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026minus;\" colname=\"c4\"\u003e \u003cp\u003e11.77% (\u0026minus;\u0026thinsp;2.85\u0026thinsp;\u0026minus;\u0026thinsp;26.39%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\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\u003eSpecificity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e85.18%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e84.91%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026minus;\" colname=\"c4\"\u003e \u003cp\u003e0.27% (\u0026minus;\u0026thinsp;10.66\u0026thinsp;\u0026minus;\u0026thinsp;11.20%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePPV\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e56.29%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e52..08%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026minus;\" colname=\"c4\"\u003e \u003cp\u003e4.21% (\u0026minus;\u0026thinsp;17.09\u0026thinsp;\u0026minus;\u0026thinsp;25.51%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.232\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNPV\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e95.77%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e92.94%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026minus;\" colname=\"c4\"\u003e \u003cp\u003e2.83% (\u0026minus;\u0026thinsp;1.73\u0026thinsp;\u0026minus;\u0026thinsp;7.39%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.042\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAUC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.920 (0.891\u0026thinsp;\u0026minus;\u0026thinsp;0.943)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.853 (0.817\u0026thinsp;\u0026minus;\u0026thinsp;0.884)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026minus;\" colname=\"c4\"\u003e \u003cp\u003e0.0244 (0.0187\u0026thinsp;\u0026minus;\u0026thinsp;0.114)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.0064\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"5\"\u003eAbbreviation: ANN, artificial neural network; PPV, positive predictive value; NPV, negative predictive value; AUC, area under the receiver operating characteristic curve.\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=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eLogistic regression analyses\u003c/h2\u003e \u003cp\u003eOne-way LR analysis indicated that a total of 24 factors were significantly associated with SAP, as shown in Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e. Among these factors, age, ADL, NEUT, LYM, hs-CRP, Alb, TG, Hb, stroke quality, COPD, AF, DOC, tracheotomy, and dysphagia showed statistically significant differences between SAP patients and non-SAP patients (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05). Multivariate LR analysis indicated that age, ADL, NEUT, LYM, hs-CRP, Alb, TG, Hb, stroke quality, COPD, AF, DOC, tracheotomy, and dysphagia were significantly associated with SAP (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05; Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Applying the LR model to the test set with a classification threshold of 0.5 yielded a SEN of 71.76% (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e) and a SPE of 84.91% .\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\u003eUnivariate analysis of factors affecting SAP following stroke\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"11\"\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 \u003cdiv align=\"left\" 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=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c11\" colnum=\"11\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"5\" nameend=\"c6\" namest=\"c2\"\u003e \u003cp\u003eUnivariate Analysis\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"5\" nameend=\"c11\" namest=\"c7\"\u003e \u003cp\u003eMultivariate Analysis\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSAP Medians (IQR)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNon-SAP Medians (IQR)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cem\u003ex\u003c/em\u003e\u003csup\u003e2\u003c/sup\u003e-Test (df)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eH-value (df)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u003cem\u003eP\u003c/em\u003e-value\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eB\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eS.E.\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eSig.\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c10\"\u003e \u003cp\u003eExp(\u003cem\u003eB\u003c/em\u003e)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c11\"\u003e \u003cp\u003e95% CI for Exp(B)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge (year)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e73 (59\u0026ndash;82)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e68 (58\u0026ndash;75)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026ndash;3.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.002\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026minus;0.011\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.013\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.408\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.989\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e0.964\u0026ndash;1.015\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGender\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e20%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e80%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.884\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.347\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFemale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e16.50%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e83.50%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eADL\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e20 (10\u0026ndash;30)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e65 (45\u0026ndash;85)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026minus;11.035\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.0001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026minus;0.054\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.009\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.947\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e0.931\u0026ndash;0.964\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWBC (\u0026times;10\u003csup\u003e9\u003c/sup\u003e/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e6.46 (5.22\u0026ndash;8.19)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e6.16 (5.19\u0026ndash;7.34)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026minus;1.511\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.131\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNEUT (\u0026times;10\u003csup\u003e9\u003c/sup\u003e/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3.87 (3.2\u0026ndash;5.43)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3.62 (2.85\u0026ndash;4.63)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026minus;-2.381\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.017\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.067\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.078\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.39\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e1.069\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e0.918\u0026ndash;1.245\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLYM (\u0026times;10\u003csup\u003e9\u003c/sup\u003e/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.54 (1.22\u0026ndash;1.81)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.72 (1.39\u0026ndash;2.155)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026minus;3.003\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.003\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026minus;0.143\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.269\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.594\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.867\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e0.512\u0026ndash;1.467\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ehs-CRP (mg/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5.94 (1.67\u0026ndash;16.33)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.75 (0.84\u0026ndash;4.365)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026minus;5.046\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.0001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026minus;0.006\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.011\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.551\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.994\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e0.973\u0026ndash;1.015\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAlb (g/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e37.3 (34\u0026ndash;40)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e41 (39\u0026ndash;43.45)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026minus;7.699\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.0001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026minus;0.137\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.006\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.872\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e0.791\u0026ndash;0.962\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTG (mmol/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.05 (0.86\u0026ndash;1.51)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.31 (1.005\u0026ndash;1.78)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026minus;2.988\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.003\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026minus;0.193\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.27\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.475\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.824\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e0.485\u0026ndash;1.400\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHb (g/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e117 (108\u0026ndash;127)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e127 (116\u0026ndash;136)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026minus;4.585\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.0001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026minus;0.005\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.011\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.672\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.995\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e0.974\u0026ndash;1.017\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eStroke quality\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCH\u0026thinsp;=\u0026thinsp;1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e28.40%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e71.60%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e6.726\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.035\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026minus;0.491\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.462\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.287\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.612\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e0.248\u0026ndash;1.512\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCI\u0026thinsp;=\u0026thinsp;2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e15.90%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e84.10%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMixed\u0026thinsp;=\u0026thinsp;3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e22.40%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e77.60%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eStroke position\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eunilateral\u0026thinsp;=\u0026thinsp;1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e16.60%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e83.40%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.477\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.294\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ebilateral\u0026thinsp;=\u0026thinsp;2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e22.90%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e77.10%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ebrainstem or cerebellum\u0026thinsp;=\u0026thinsp;3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e21.80%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e78.20%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCOPD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\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\u003e54.50%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e45.50%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e9.582\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.002\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e2.744\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e1.078\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.011\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e15.551\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e1.879\u0026ndash;128.737\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\u003e17.80%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e82.20%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\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\u003e13.90%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e86.10%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3.546\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\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\u003e21.10%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e78.90%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHypertension\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\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\u003e18.80%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e81.20%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.015\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.904\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e18.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e81.70%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAF\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\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\u003e52.50%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e47.50%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e33.146\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.0001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e1.32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.529\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.013\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e3.744\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e1.327\u0026ndash;10.564\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\u003e15.40%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e84.60%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSmoking history\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\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\u003e18.80%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e81.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.987\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e18.60%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e81.40%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003edrinking history\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\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\u003e19.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e81.00%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.002\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.961\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e18.60%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e81.40%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDOC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\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\u003e65.70%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e34.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e55.392\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.0001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.118\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.598\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.843\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e1.125\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e0.349\u0026ndash;3.632\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\u003e14.70%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e85.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eTracheotomy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\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\u003e95.80%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4.20%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e99.537\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.0001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e2.763\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e1.111\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.013\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e15.855\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e1.796\u0026ndash;139.997\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\u003e14.40%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e85.60%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNasal feeding tube\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\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\u003e80.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e19.70%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e210.69\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.0001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e7.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e92.70%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDysphagia\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\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\u003e35.10%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e64.90%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e16.479\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.0001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.36\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.386\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.351\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e1.434\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e0.673\u0026ndash;3.057\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\u003e15.30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e84.70%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"11\"\u003eADL: activities of daily living; WBC: white cell count; NEUT: neutrophil count; LYM: lymphocyte count; hs-CRP: hypersensitive C-reactive protein; Alb: albumin; TG: triglyceride; Hb: hemoglobin; CH: cerebral hemorrhage CI: cerebral infarction; COPD: chronic obstructive pulmonary disease; DM: diabetes mellitus; AF: atrial fibrillation; DOC: disorders of consciousness\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003eComparison of the LR and ANN models\u003c/h2\u003e \u003cp\u003eThe prediction results of both the LR and ANN models are summarized in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e. A significant difference in SEN was observed between the models (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001). The test dataset used to identify SAP was also employed to construct the ROC curves for both models. The AUC values of the LR and ANN models were 0.853 (0.817\u0026ndash;0.884) and 0.920 (0.891\u0026ndash;0.943), respectively, indicating that the ANN model was more accurate overall (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e; \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0064).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eSAP is an important determinant of adverse clinical outcomes and mortality after stroke.\u003csup\u003e[\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]\u003c/sup\u003e The mortality rate of stroke patients with pneumonia is 5.58 times that of stroke patients without pneumonia.\u003csup\u003e[\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]\u003c/sup\u003e Infection is a risk factor for stroke and plays an important role in the prognosis after stroke.\u003csup\u003e[\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]\u003c/sup\u003e Infection, including SAP, is a risk factor for stroke and plays a crucial role in post-stroke prognosis. Due to its higher prevalence and challenging treatment, SAP has a more significant influence on clinical outcomes.\u003csup\u003e[\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]\u003c/sup\u003e At present, the assessment of SAP risk remains challenging in clinical practice. Therefore, it is necessary to identify patients at risk for SAP at an early stage, which may facilitate the implementation of monitoring and preventive measures in routine clinical care. In this study, we compared ANN and LR models for identifying stroke patients at risk of SAP.\u003c/p\u003e \u003cp\u003eANNs, which do not suffer from the uncertainties inherent in linearized models, can be used to analyze complex nonlinear relationships in the medical field with the goal of improving system performance.\u003csup\u003e[\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]\u003c/sup\u003e Consequently, ANN models are favored over LR models because of their superior predictive accuracy when addressing complex nonlinear problems.\u003csup\u003e[\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]\u003c/sup\u003e ANNs have shown promise in numerous applications including biochemical analysis, disease prediction, and diagnostic systems.\u003csup\u003e[\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eIn the current study, the ANN model exhibited greater accuracy in predicting the occurrence of SAP compared to the LR model. The ANN model for predicting SAP outperformed the LR approach in terms of SEN, SPE, PPV, and NPV.\u003c/p\u003e \u003cp\u003eWe also identified several predictors of SAP, including dysphagia and C-reactive protein, in alignment with previous studies suggesting that these factors are associated with the incidence of SAP.\u003csup\u003e[\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]\u003c/sup\u003e We incorporated additional variables into the ANN model to enhance its predictive capacity. Notably, we found a significant association between Alb and SAP. Albumin, which has been shown to have a high diagnostic value, is significantly linked to the development of SAP in older patients; the risk of SAP increases as the albumin level declines.\u003csup\u003e[\u003cspan additionalcitationids=\"CR34\" citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]\u003c/sup\u003e Albumin has an antioxidant effect; thus, a reduction in Alb may lead to increased oxidative stress, increased cell damage and inflammation, and a weakening of the blood\u0026ndash;brain barrier, making it easier for infectious agents to enter and increasing the risk of secondary infection after stroke.\u003csup\u003e[\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]\u003c/sup\u003e Additionally, we identified a lower ADL score as an independent risk factor for pneumonia. A low ADL score typically indicates a high level of nursing dependence; among stroke patients, those with relatively low ADL scores upon hospital admission were found to have a higher likelihood of developing SAP.\u003csup\u003e[\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e]\u003c/sup\u003e In the present study, we also showed that low hemoglobin levels were an independent risk factor for SAP. Low hemoglobin concentrations are associated with a reduced capacity for oxygen transport, leading to an increased likelihood of infarction in regions with compromised blood flow.Decreased hemoglobin concentrations are associated with a diminished capacity for oxygen transport, resulting in increased likelihood of infarction in regions with compromised blood flow.\u003csup\u003e[\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e]\u003c/sup\u003e Stroke patients may be nutritionally imbalanced, further increasing the rate of infection.\u003csup\u003e[\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e]\u003c/sup\u003e CRP has the potential to predict the occurrence of SAP\u003csup\u003e[\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e]\u003c/sup\u003e because elevated CRP levels are closely linked to the progression of SAP.\u003csup\u003e[\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e]\u003c/sup\u003e The levels of acute-phase plasma proteins such as CRP increase significantly during the acute phase of an inflammatory reaction.\u003csup\u003e[\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]\u003c/sup\u003e When CRP levels are elevated, it reflects the activation of the inflammatory response in the body, which may lead to the growth of pathogens such as bacteria and viruses in the lungs, which can lead to pneumonia. In stroke patients, the inflammatory response is exacerbated by factors such as weakened immunity and impaired coughing and swallowing reflexes, which further increases the risk of SAP\u003csup\u003e[\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eIt is important to recognize the constraints inherent in the present study. First, this was a single-center retrospective study, which inherently restricts the generalizability of the findings. The presence of missing values in some analyzed variables also limits the utility of the findings. Second, our findings may be subject to potential bias due to insufficient available data and lack of external validation. Consequently, additional multicenter, prospective studies are needed to overcome these limitations.\u003c/p\u003e"},{"header":"Conclusions","content":"\u003cp\u003eWe successfully developed and validated an ANN model to identify stroke patients at risk for SAP during their preliminary evaluation or at the point of entry into the emergency department. The application of this predictive model promises significant improvements in the early identification of SAP cases, enabling timely interventions that could mitigate the progression of SAP. Future efforts will focus on further optimizing the model to enhance its clinical applicability in diverse healthcare settings.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003eFunding\u003c/p\u003e\n\u003cp\u003eThis work was supported by the Natural Key Projects of Research Projects in Anhui Province Universities (2023AH050784).\u003c/p\u003e\n\u003cp\u003eAuthor contributions\u003c/p\u003e\n\u003cp\u003eTing Wang, Linli Yuan, Min You, Juan Yuan and Sijing Peng conceived of and designed the study;Ting Wang, Chunbiao Li conducted the experiments and authored the manuscript; Ting Wang, Chunbiao Liworked on the data integration,preprocessing and analysis; Yi Liu, Min Yang, Yaling Fan, Qinsi Tong and\u0026nbsp;Dajin Li\u0026nbsp;collected the data;Min You, Juan Yuan\u0026nbsp;oversaw the project, coordinated;All authors read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003eConflicts of interest\u003c/p\u003e\n\u003cp\u003eThere are no conflicts of interest to be declared.\u003c/p\u003e\n\u003cp\u003eEthics approval and consent to participate\u003c/p\u003e\n\u003cp\u003eThis study was approved by the Ethics Committee of the Second Affiliated Hospital of Anhui University of Traditional Chinese Medicine (2021-zj-24). In addition, informed consent was omitted for all participants as the study was retrospective in nature, and the research process was in accordance with the Declaration of Helsinki.\u003c/p\u003e\n\u003cp\u003eAcknowledgments\u003c/p\u003e\n\u003cp\u003eThe authors thank AiMi Academic Services (www.aimieditor.com) for English language editing and review services.\u003c/p\u003e\n\u003cp\u003eDeclaration of Generative AI and AI-assisted technologies in the writing process\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eSmith CJ, Kishore AK, Vail A, et al. Diagnosis of Stroke-Associated Pneumonia: Recommendations from the Pneumonia in Stroke Consensus Group. Stroke. 2015;46:2335\u0026ndash;40.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWestendorp WF, Nederkoorn PJ, Vermeij J-D, et al. Post-stroke infection: a systematic review and meta-analysis. BMC Neurol. 2011;11:110.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBadve MS, Zhou Z, van de Beek D, et al. Frequency of post-stroke pneumonia: Systematic review and meta-analysis of observational studies. Int J Stroke Off J Int Stroke Soc. 2019;14:125\u0026ndash;36.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTeh WH, Smith CJ, Barlas RS, et al. Impact of stroke-associated pneumonia on mortality, length of hospitalization, and functional outcome. Acta Neurol Scand. 2018;138:293\u0026ndash;300.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAli AN, Howe J, Majid A, et al. The economic cost of stroke-associated pneumonia in a UK setting. Top Stroke Rehabil. 2018;25:214\u0026ndash;23.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKwon H-M, Jeong S-W, Lee S-H, et al. The pneumonia score: A simple grading scale for prediction of pneumonia after acute stroke. Am J Infect Control. 2006;34:64\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHoffmann S, Malzahn U, Harms H, et al. Development of a Clinical Score (A2DS2) to Predict Pneumonia in Acute Ischemic Stroke. Stroke. 2012;43:2617\u0026ndash;23.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSmith CJ, Bray BD, Hoffman A, et al. Can a Novel Clinical Risk Score Improve Pneumonia Prediction in Acute Stroke Care? A UK Multicenter Cohort Study. J Am Heart Assoc. 2015;4:e001307.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKumar S, Marchina S, Massaro J, et al. ACDD4 score: A simple tool for assessing risk of pneumonia after stroke. J Neurol Sci. 2017;372:399\u0026ndash;402.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuang G-Q, Lin Y-T, Wu Y-M et al. Individualized Prediction Of Stroke-Associated Pneumonia For Patients With Acute Ischemic Stroke. CIA. 2019; Volume 14:1951\u0026ndash;1962.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJiang F, Dong L, Dai Q. Electrical resistivity imaging inversion: An ISFLA trained kernel principal component wavelet neural network approach. Neural Netw Off J Int Neural Netw Soc. 2018;104:114\u0026ndash;23.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFei Y, Gao K, Li W-Q. Artificial neural network algorithm model as powerful tool to predict acute lung injury following to severe acute pancreatitis. Pancreatol Off J Int Assoc Pancreatol IAP Al. 2018;18:892\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChung C-C, Chan L, Bamodu OA, et al. Artificial neural network based prediction of postthrombolysis intracerebral hemorrhage and death. Sci Rep. 2020;10:20501.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHou J, Fu S, Wang X, et al. A noninvasive artificial neural network model to predict IgA nephropathy risk in Chinese population. Sci Rep. 2022;12:8296.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTafeit E, Reibnegger G. Artificial neural networks in laboratory medicine and medical outcome prediction. Clin Chem Lab Med. 1999;37:845\u0026ndash;53.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNaidu SMM, Pandey PC, Bagal UR, et al. Beat-to-beat estimation of stroke volume using impedance cardiography and artificial neural network. Med Biol Eng Comput. 2018;56:1077\u0026ndash;89.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMoon S, Ahmadnezhad P, Song H-J, et al. Artificial neural networks in neurorehabilitation: A scoping review. NeuroRehabilitation. 2020;46:259\u0026ndash;69.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLu F, Liang Y, Wang X, et al. Prediction of amorphous forming ability based on artificial neural network and convolutional neural network. Comput Mater Sci. 2022;210:111464.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFei Y, Gao K, Li W. Prediction and evaluation of the severity of acute respiratory distress syndrome following severe acute pancreatitis using an artificial neural network algorithm model. HPB. 2019;21:891\u0026ndash;7.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFei Y, Gao K, Li W. Artificial neural network algorithm model as powerful tool to predict acute lung injury following to severe acute pancreatitis. Pancreatology. 2018;18:892\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShariatnia S, Ziaratban M, Rajabi A, et al. Modeling the diagnosis of coronary artery disease by discriminant analysis and logistic regression: a cross-sectional study. BMC Med Inf Decis Mak. 2022;22:85.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePatel UK, Kodumuri N, Dave M, et al. Stroke-Associated Pneumonia: A Retrospective Study of Risk Factors and Outcomes. Neurologist. 2020;25:39\u0026ndash;48.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWestendorp WF, Nederkoorn PJ, Vermeij J-D, et al. Post-stroke infection: A systematic review and meta-analysis. BMC Neurol. 2011;11:110.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eElkind MSV, Boehme AK, Smith CJ, et al. Infection as a Stroke Risk Factor and Determinant of Outcome After Stroke. Stroke. 2020;51:3156\u0026ndash;68.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHilker R, Poetter C, Findeisen N, et al. Nosocomial Pneumonia After Acute Stroke: Implications for Neurological Intensive Care Medicine. Stroke. 2003;34:975\u0026ndash;81.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHu J, Liu L, Wang Y, et al. Precision motion control of a small launching platform with disturbance compensation using neural networks: INTELLIGENT CONTROL IN LAUNCHING TECHNOLOGY. Int J Adapt Control Signal Process. 2017;31:971\u0026ndash;84.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOttenbacher KJ, Linn RT, Smith PM, et al. Comparison of logistic regression and neural network analysis applied to predicting living setting after hip fracture. Ann Epidemiol. 2004;14:551\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDreiseitl S, Ohno-Machado L. Logistic regression and artificial neural network classification models: a methodology review. J Biomed Inf. 2002;35:352\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChun-An Cheng null. Hung-Wen Chiu null. An artificial neural network model for the evaluation of carotid artery stenting prognosis using a national-wide database. Annu Int Conf IEEE Eng Med Biol Soc IEEE Eng Med Biol Soc Annu Int Conf 2017; 2017:2566\u0026ndash;2569.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRau H-H, Hsu C-Y, Lin Y-A, et al. Development of a web-based liver cancer prediction model for type II diabetes patients by using an artificial neural network. Comput Methods Programs Biomed. 2016;125:58\u0026ndash;65.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuang G-Q, Lin Y-T, Wu Y-M, et al. Individualized Prediction Of Stroke-Associated Pneumonia For Patients With Acute Ischemic Stroke. Clin Interv Aging. 2019;14:1951\u0026ndash;62.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi X, Wu M, Sun C, et al. Using machine learning to predict stroke-associated pneumonia in Chinese acute ischaemic stroke patients. Eur J Neurol. 2020;27:1656\u0026ndash;63.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWiedermann CJ. Hypoalbuminemia as Surrogate and Culprit of Infections. Int J Mol Sci. 2021;22:4496.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLv X-N, Shen Y-Q, Li Z-Q, et al. Neutrophil percentage to albumin ratio is associated with stroke-associated pneumonia and poor outcome in patients with spontaneous intracerebral hemorrhage. Front Immunol. 2023;14:1173718.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZawiah M, Khan AH, Abu Farha R, et al. Predictors of stroke-associated pneumonia and the predictive value of neutrophil percentage-to-albumin ratio. Postgrad Med. 2023;135:681\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHalliwell B. Albumin\u0026mdash;An important extracellular antioxidant? Biochem Pharmacol. 1988;37:569\u0026ndash;71.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWatanabe S, Shimozato K, Oh-Shige H, et al. Examination of factors associated with aspiration pneumonia following stroke. Oral Sci Int. 2014;11:15\u0026ndash;21.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNaess H, Logallo N, Waje-Andreassen U, et al. U‐shaped relationship between hemoglobin level and severity of ischemic stroke. Acta Neurol Scand. 2019;140:56\u0026ndash;61.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSong X, He Y, Bai J, et al. A nomogram based on nutritional status and A2DS2 score for predicting stroke-associated pneumonia in acute ischemic stroke patients with type 2 diabetes mellitus: A retrospective study. Front Nutr. 2022;9:1009041.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi Y, Zhao L, Liu Y, et al. Novel Predictors of Stroke-Associated Pneumonia: A Single Center Analysis. Front Neurol. 2022;13:857420.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKalra L, Smith CJ, Hodsoll J, et al. Elevated C-reactive protein increases diagnostic accuracy of algorithm-defined stroke-associated pneumonia in afebrile patients. Int J Stroke. 2019;14:167\u0026ndash;73.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKalra L, Smith CJ, Hodsoll J, et al. Elevated C-reactive protein increases diagnostic accuracy of algorithm-defined stroke-associated pneumonia in afebrile patients. Int J Stroke. 2019;14:167\u0026ndash;73.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"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":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Stroke, Stroke-associated pneumonia, Artificial neural networks, risk, prediction","lastPublishedDoi":"10.21203/rs.3.rs-4754561/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4754561/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eObjective\u003c/h2\u003e \u003cp\u003eThis study developed a predictive model for the risk of stroke-associated pneumonia (SAP) based on an advanced artificial neural network (ANN) model.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eUsing clinical and laboratory data collected from a cohort of 456 hospital patients from July 2017 to April 2022, we constructed ANN and logistic regression (LR) models. The models were trained on a randomly selected group of 292 patients, and subsequent model validation and testing were carried out on two separate sets of 82 patients each. The predictive performances of both models were evaluated using a comprehensive range of statistical indices.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eDuring dataset partitioning, the 24 variables across the training, validation, and test sets displayed no significant discrepancies. The prediction performance of the ANN model was better than that of the LR model. When applied to the test cohort, the ANN model had a sensitivity of 83.53% and a specificity of 85.18%. Comparative analysis revealed discernible discrepancies between the performance indexes of the ANN and LR models. Based on the receiver operating characteristic curve, the ANN model showed robust ability to identify SAP, with an area under the curve value of 0.920. The principal independent predictors in the model were serum albumin, activities of daily living score, hemoglobin level, and hypersensitive C-reactive protein level.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eThe developed ANN model demonstrates promising predictive capability for assessing the risk of SAP. However, further verification with larger and more diverse datasets is needed to confirm its utility as a tool for clinical prediction.\u003c/p\u003e","manuscriptTitle":"Prediction and evaluation of the risk of Stroke-associated pneumonia using an artificial neural network model","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-08-18 23:46:25","doi":"10.21203/rs.3.rs-4754561/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"286e53f8-50aa-4197-bdc6-3acd727cb0d8","owner":[],"postedDate":"August 18th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-01-22T08:39:06+00:00","versionOfRecord":[],"versionCreatedAt":"2024-08-18 23:46:25","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4754561","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4754561","identity":"rs-4754561","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}