Development and validation of an early predictive model for hemiplegic shoulder pain: a comparative study of logistic regression, support vector machine, and random forest

ObjectiveIn this study, we aim to identify the predictive variables for hemiplegic shoulder pain (HSP) through machine learning algorithms, select the optimal model and predict the occurrence of HSP.MethodsData of 332 stroke patients admitted to a tertiary hospital in Zhejiang Province from January...

Full description

Saved in:
Bibliographic Details
Main Authors: Qiang Wu, Fang Zhang, Yuchang Fei, Zhenfen Sima, Shanshan Gong, Qifeng Tong, Qingchuan Jiao, Hao Wu, Jianqiu Gong
Format: Article
Language:English
Published: Frontiers Media S.A. 2025-06-01
Series:Frontiers in Neurology
Subjects:
Online Access:https://www.frontiersin.org/articles/10.3389/fneur.2025.1612222/full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849422315833524224
author Qiang Wu
Qiang Wu
Fang Zhang
Yuchang Fei
Zhenfen Sima
Shanshan Gong
Qifeng Tong
Qingchuan Jiao
Hao Wu
Jianqiu Gong
Jianqiu Gong
author_facet Qiang Wu
Qiang Wu
Fang Zhang
Yuchang Fei
Zhenfen Sima
Shanshan Gong
Qifeng Tong
Qingchuan Jiao
Hao Wu
Jianqiu Gong
Jianqiu Gong
author_sort Qiang Wu
collection DOAJ
description ObjectiveIn this study, we aim to identify the predictive variables for hemiplegic shoulder pain (HSP) through machine learning algorithms, select the optimal model and predict the occurrence of HSP.MethodsData of 332 stroke patients admitted to a tertiary hospital in Zhejiang Province from January 2022 to January 2023 were collected. After screening predictive variables by LASSO regression, three predictive models selected using the LazyPredict package, namely logistic regression (LR), support vector machine (SVM) and random forest (RF), were established respectively. The performance parameters (accuracy, precision, recall, and F1 score) of the models were calculated, the receiver operating characteristic curve (ROC) and the decision curve analysis (DCA) were plotted to compare the performance of the three models. An explainability analysis (SHAP) was conducted on the optimal model.ResultsThe RF model performed the best, with accuracy: 0.90, precision: 0.89, recall: 0.88, F1 score: 0.86, AUC-ROC: 0.94, and the range of the threshold probability in DCA: 7%−99%. Based on the SHAP analysis of the explainability of the RF model, the contribution degrees of the early HSP predictive variables from high to low are as follows: multiple injuries, shoulder joint flexion (p), biceps tendon effusion, sensory disorder, supraspinatus tendinopathy, subluxation, diabetes, and age.ConclusionThe RF prediction model has a good predictive effect on HSP and has good clinical explainability. It can provide objective references for the early warning and stratified management of HSP.
format Article
id doaj-art-3a9e4ddaa1b247a0bb15e7123d2fb0ee
institution Kabale University
issn 1664-2295
language English
publishDate 2025-06-01
publisher Frontiers Media S.A.
record_format Article
series Frontiers in Neurology
spelling doaj-art-3a9e4ddaa1b247a0bb15e7123d2fb0ee2025-08-20T03:31:10ZengFrontiers Media S.A.Frontiers in Neurology1664-22952025-06-011610.3389/fneur.2025.16122221612222Development and validation of an early predictive model for hemiplegic shoulder pain: a comparative study of logistic regression, support vector machine, and random forestQiang Wu0Qiang Wu1Fang Zhang2Yuchang Fei3Zhenfen Sima4Shanshan Gong5Qifeng Tong6Qingchuan Jiao7Hao Wu8Jianqiu Gong9Jianqiu Gong10Department of Rehabilitation Medicine, The First Affiliated Hospital, Shaoxing University, Shaoxing, Zhejiang, ChinaSchool of Medicine, Shaoxing University, Shaoxing, Zhejiang, ChinaDepartment of Rehabilitation Medicine, The First Affiliated Hospital, Shaoxing University, Shaoxing, Zhejiang, ChinaDepartment of Integrated Chinese and Western Medicine, The First People's Hospital of Jiashan, Jiaxing, Zhejiang, ChinaDepartment of Rehabilitation Medicine, The First Affiliated Hospital, Shaoxing University, Shaoxing, Zhejiang, ChinaDepartment of Gastroenterology, The Third Affiliated Hospital of Zhejiang Chinese Medical University, Hangzhou, Zhejiang, ChinaDepartment of Rehabilitation Medicine, The First Affiliated Hospital, Shaoxing University, Shaoxing, Zhejiang, ChinaDepartment of Rehabilitation Medicine, The First Affiliated Hospital, Shaoxing University, Shaoxing, Zhejiang, ChinaDepartment of Rehabilitation Medicine, The First Affiliated Hospital, Shaoxing University, Shaoxing, Zhejiang, ChinaDepartment of Rehabilitation Medicine, The First Affiliated Hospital, Shaoxing University, Shaoxing, Zhejiang, ChinaSchool of Medicine, Shaoxing University, Shaoxing, Zhejiang, ChinaObjectiveIn this study, we aim to identify the predictive variables for hemiplegic shoulder pain (HSP) through machine learning algorithms, select the optimal model and predict the occurrence of HSP.MethodsData of 332 stroke patients admitted to a tertiary hospital in Zhejiang Province from January 2022 to January 2023 were collected. After screening predictive variables by LASSO regression, three predictive models selected using the LazyPredict package, namely logistic regression (LR), support vector machine (SVM) and random forest (RF), were established respectively. The performance parameters (accuracy, precision, recall, and F1 score) of the models were calculated, the receiver operating characteristic curve (ROC) and the decision curve analysis (DCA) were plotted to compare the performance of the three models. An explainability analysis (SHAP) was conducted on the optimal model.ResultsThe RF model performed the best, with accuracy: 0.90, precision: 0.89, recall: 0.88, F1 score: 0.86, AUC-ROC: 0.94, and the range of the threshold probability in DCA: 7%−99%. Based on the SHAP analysis of the explainability of the RF model, the contribution degrees of the early HSP predictive variables from high to low are as follows: multiple injuries, shoulder joint flexion (p), biceps tendon effusion, sensory disorder, supraspinatus tendinopathy, subluxation, diabetes, and age.ConclusionThe RF prediction model has a good predictive effect on HSP and has good clinical explainability. It can provide objective references for the early warning and stratified management of HSP.https://www.frontiersin.org/articles/10.3389/fneur.2025.1612222/fullhemiplegic shoulder painprediction modelrandom forestsupport vector machineSHAP
spellingShingle Qiang Wu
Qiang Wu
Fang Zhang
Yuchang Fei
Zhenfen Sima
Shanshan Gong
Qifeng Tong
Qingchuan Jiao
Hao Wu
Jianqiu Gong
Jianqiu Gong
Development and validation of an early predictive model for hemiplegic shoulder pain: a comparative study of logistic regression, support vector machine, and random forest
Frontiers in Neurology
hemiplegic shoulder pain
prediction model
random forest
support vector machine
SHAP
title Development and validation of an early predictive model for hemiplegic shoulder pain: a comparative study of logistic regression, support vector machine, and random forest
title_full Development and validation of an early predictive model for hemiplegic shoulder pain: a comparative study of logistic regression, support vector machine, and random forest
title_fullStr Development and validation of an early predictive model for hemiplegic shoulder pain: a comparative study of logistic regression, support vector machine, and random forest
title_full_unstemmed Development and validation of an early predictive model for hemiplegic shoulder pain: a comparative study of logistic regression, support vector machine, and random forest
title_short Development and validation of an early predictive model for hemiplegic shoulder pain: a comparative study of logistic regression, support vector machine, and random forest
title_sort development and validation of an early predictive model for hemiplegic shoulder pain a comparative study of logistic regression support vector machine and random forest
topic hemiplegic shoulder pain
prediction model
random forest
support vector machine
SHAP
url https://www.frontiersin.org/articles/10.3389/fneur.2025.1612222/full
work_keys_str_mv AT qiangwu developmentandvalidationofanearlypredictivemodelforhemiplegicshoulderpainacomparativestudyoflogisticregressionsupportvectormachineandrandomforest
AT qiangwu developmentandvalidationofanearlypredictivemodelforhemiplegicshoulderpainacomparativestudyoflogisticregressionsupportvectormachineandrandomforest
AT fangzhang developmentandvalidationofanearlypredictivemodelforhemiplegicshoulderpainacomparativestudyoflogisticregressionsupportvectormachineandrandomforest
AT yuchangfei developmentandvalidationofanearlypredictivemodelforhemiplegicshoulderpainacomparativestudyoflogisticregressionsupportvectormachineandrandomforest
AT zhenfensima developmentandvalidationofanearlypredictivemodelforhemiplegicshoulderpainacomparativestudyoflogisticregressionsupportvectormachineandrandomforest
AT shanshangong developmentandvalidationofanearlypredictivemodelforhemiplegicshoulderpainacomparativestudyoflogisticregressionsupportvectormachineandrandomforest
AT qifengtong developmentandvalidationofanearlypredictivemodelforhemiplegicshoulderpainacomparativestudyoflogisticregressionsupportvectormachineandrandomforest
AT qingchuanjiao developmentandvalidationofanearlypredictivemodelforhemiplegicshoulderpainacomparativestudyoflogisticregressionsupportvectormachineandrandomforest
AT haowu developmentandvalidationofanearlypredictivemodelforhemiplegicshoulderpainacomparativestudyoflogisticregressionsupportvectormachineandrandomforest
AT jianqiugong developmentandvalidationofanearlypredictivemodelforhemiplegicshoulderpainacomparativestudyoflogisticregressionsupportvectormachineandrandomforest
AT jianqiugong developmentandvalidationofanearlypredictivemodelforhemiplegicshoulderpainacomparativestudyoflogisticregressionsupportvectormachineandrandomforest