A clinical prediction model based on interpretable machine learning algorithms for prolonged hospital stay in acute ischemic stroke patients: a real-world study
ObjectiveAcute ischemic stroke (AIS) brings an increasingly heavier economic burden nowadays. Prolonged length of stay (LOS) is a vital factor in healthcare expenditures. The aim of this study was to predict prolonged LOS in AIS patients based on an interpretable machine learning algorithm.MethodsWe...
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2023-11-01
|
| Series: | Frontiers in Endocrinology |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fendo.2023.1165178/full |
| _version_ | 1797488429617905664 |
|---|---|
| author | Kai Wang Kai Wang Qianmei Jiang Murong Gao Xiu’e Wei Xiu’e Wei Chan Xu Chengliang Yin Haiyan Liu Haiyan Liu Renjun Gu Haosheng Wang Haosheng Wang Wenle Li Wenle Li Liangqun Rong Liangqun Rong |
| author_facet | Kai Wang Kai Wang Qianmei Jiang Murong Gao Xiu’e Wei Xiu’e Wei Chan Xu Chengliang Yin Haiyan Liu Haiyan Liu Renjun Gu Haosheng Wang Haosheng Wang Wenle Li Wenle Li Liangqun Rong Liangqun Rong |
| author_sort | Kai Wang |
| collection | DOAJ |
| description | ObjectiveAcute ischemic stroke (AIS) brings an increasingly heavier economic burden nowadays. Prolonged length of stay (LOS) is a vital factor in healthcare expenditures. The aim of this study was to predict prolonged LOS in AIS patients based on an interpretable machine learning algorithm.MethodsWe enrolled AIS patients in our hospital from August 2017 to July 2019, and divided them into the “prolonged LOS” group and the “no prolonged LOS” group. Prolonged LOS was defined as hospitalization for more than 7 days. The least absolute shrinkage and selection operator (LASSO) regression was applied to reduce the dimensionality of the data. We compared the predictive capacity of extended LOS in eight different machine learning algorithms. SHapley Additive exPlanations (SHAP) values were used to interpret the outcome, and the most optimal model was assessed by discrimination, calibration, and clinical utility.ResultsProlonged LOS developed in 149 (22.0%) of the 677 eligible patients. In eight machine learning algorithms, prolonged LOS was best predicted by the Gaussian naive Bayes (GNB) model, which had a striking area under the curve (AUC) of 0.878 ± 0.007 in the training set and 0.857 ± 0.039 in the validation set. The variables sorted by the gap values showed that the strongest predictors were pneumonia, dysphagia, thrombectomy, and stroke severity. High net benefits were observed at 0%–76% threshold probabilities, while good agreement was found between the observed and predicted probabilities.ConclusionsThe model using the GNB algorithm proved excellent for predicting prolonged LOS in AIS patients. This simple model of prolonged hospitalization could help adjust policies and better utilize resources. |
| first_indexed | 2024-03-10T00:03:01Z |
| format | Article |
| id | doaj.art-2c9ed40df3a247518e8e04e91a142120 |
| institution | Directory Open Access Journal |
| issn | 1664-2392 |
| language | English |
| last_indexed | 2024-03-10T00:03:01Z |
| publishDate | 2023-11-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Endocrinology |
| spelling | doaj.art-2c9ed40df3a247518e8e04e91a1421202023-11-23T16:12:57ZengFrontiers Media S.A.Frontiers in Endocrinology1664-23922023-11-011410.3389/fendo.2023.11651781165178A clinical prediction model based on interpretable machine learning algorithms for prolonged hospital stay in acute ischemic stroke patients: a real-world studyKai Wang0Kai Wang1Qianmei Jiang2Murong Gao3Xiu’e Wei4Xiu’e Wei5Chan Xu6Chengliang Yin7Haiyan Liu8Haiyan Liu9Renjun Gu10Haosheng Wang11Haosheng Wang12Wenle Li13Wenle Li14Liangqun Rong15Liangqun Rong16Department of Neurology, The Second Affiliated Hospital of Xuzhou Medical University, Xuzhou, Jiangsu, ChinaKey Laboratory of Neurological Diseases, The Second Affiliated Hospital of Xuzhou Medical University, Xuzhou, Jiangsu, ChinaDepartment of General Practice, Xindu District People’s Hospital of Chengdu, Chengdu, Sichuan, ChinaDepartment of Rehabilitation, Beijing Rehabilitation Hospital Affiliated to Capital Medical University, Beijing, ChinaDepartment of Neurology, The Second Affiliated Hospital of Xuzhou Medical University, Xuzhou, Jiangsu, ChinaKey Laboratory of Neurological Diseases, The Second Affiliated Hospital of Xuzhou Medical University, Xuzhou, Jiangsu, ChinaDepartment of Dermatology, Xianyang Central Hospital, Xianyang, ChinaFaculty of Medicine, Macau University of Science and Technology, Macau, Macao SAR, ChinaDepartment of Neurology, The Second Affiliated Hospital of Xuzhou Medical University, Xuzhou, Jiangsu, ChinaKey Laboratory of Neurological Diseases, The Second Affiliated Hospital of Xuzhou Medical University, Xuzhou, Jiangsu, ChinaSchool of Chinese Medicine and School of Integrated Chinese and Western Medicine, Nanjing University of Chinese Medicine, Nanjing, ChinaSchool of Chinese Medicine and School of Integrated Chinese and Western Medicine, Nanjing University of Chinese Medicine, Nanjing, ChinaState Key Laboratory of Pharmaceutical Biotechnology, Division of Sports Medicine and Adult Reconstructive Surgery, Department of Orthopedic Surgery, Nanjing Drum Tower Hospital, The Affiliated Hospital of Nanjing University Medical School, Nanjing, Jiangsu, ChinaKey Laboratory of Neurological Diseases, The Second Affiliated Hospital of Xuzhou Medical University, Xuzhou, Jiangsu, ChinaThe State Key Laboratory of Molecular Vaccinology and Molecular Diagnostics and Center for Molecular Imaging and Translational Medicine, School of Public Health, Xiamen University, Xiamen, ChinaDepartment of Neurology, The Second Affiliated Hospital of Xuzhou Medical University, Xuzhou, Jiangsu, ChinaKey Laboratory of Neurological Diseases, The Second Affiliated Hospital of Xuzhou Medical University, Xuzhou, Jiangsu, ChinaObjectiveAcute ischemic stroke (AIS) brings an increasingly heavier economic burden nowadays. Prolonged length of stay (LOS) is a vital factor in healthcare expenditures. The aim of this study was to predict prolonged LOS in AIS patients based on an interpretable machine learning algorithm.MethodsWe enrolled AIS patients in our hospital from August 2017 to July 2019, and divided them into the “prolonged LOS” group and the “no prolonged LOS” group. Prolonged LOS was defined as hospitalization for more than 7 days. The least absolute shrinkage and selection operator (LASSO) regression was applied to reduce the dimensionality of the data. We compared the predictive capacity of extended LOS in eight different machine learning algorithms. SHapley Additive exPlanations (SHAP) values were used to interpret the outcome, and the most optimal model was assessed by discrimination, calibration, and clinical utility.ResultsProlonged LOS developed in 149 (22.0%) of the 677 eligible patients. In eight machine learning algorithms, prolonged LOS was best predicted by the Gaussian naive Bayes (GNB) model, which had a striking area under the curve (AUC) of 0.878 ± 0.007 in the training set and 0.857 ± 0.039 in the validation set. The variables sorted by the gap values showed that the strongest predictors were pneumonia, dysphagia, thrombectomy, and stroke severity. High net benefits were observed at 0%–76% threshold probabilities, while good agreement was found between the observed and predicted probabilities.ConclusionsThe model using the GNB algorithm proved excellent for predicting prolonged LOS in AIS patients. This simple model of prolonged hospitalization could help adjust policies and better utilize resources.https://www.frontiersin.org/articles/10.3389/fendo.2023.1165178/fullprolonged hospital staystrokemachine learningprediction modelSHAP (SHapley Additive exPlanations) |
| spellingShingle | Kai Wang Kai Wang Qianmei Jiang Murong Gao Xiu’e Wei Xiu’e Wei Chan Xu Chengliang Yin Haiyan Liu Haiyan Liu Renjun Gu Haosheng Wang Haosheng Wang Wenle Li Wenle Li Liangqun Rong Liangqun Rong A clinical prediction model based on interpretable machine learning algorithms for prolonged hospital stay in acute ischemic stroke patients: a real-world study Frontiers in Endocrinology prolonged hospital stay stroke machine learning prediction model SHAP (SHapley Additive exPlanations) |
| title | A clinical prediction model based on interpretable machine learning algorithms for prolonged hospital stay in acute ischemic stroke patients: a real-world study |
| title_full | A clinical prediction model based on interpretable machine learning algorithms for prolonged hospital stay in acute ischemic stroke patients: a real-world study |
| title_fullStr | A clinical prediction model based on interpretable machine learning algorithms for prolonged hospital stay in acute ischemic stroke patients: a real-world study |
| title_full_unstemmed | A clinical prediction model based on interpretable machine learning algorithms for prolonged hospital stay in acute ischemic stroke patients: a real-world study |
| title_short | A clinical prediction model based on interpretable machine learning algorithms for prolonged hospital stay in acute ischemic stroke patients: a real-world study |
| title_sort | clinical prediction model based on interpretable machine learning algorithms for prolonged hospital stay in acute ischemic stroke patients a real world study |
| topic | prolonged hospital stay stroke machine learning prediction model SHAP (SHapley Additive exPlanations) |
| url | https://www.frontiersin.org/articles/10.3389/fendo.2023.1165178/full |
| work_keys_str_mv | AT kaiwang aclinicalpredictionmodelbasedoninterpretablemachinelearningalgorithmsforprolongedhospitalstayinacuteischemicstrokepatientsarealworldstudy AT kaiwang aclinicalpredictionmodelbasedoninterpretablemachinelearningalgorithmsforprolongedhospitalstayinacuteischemicstrokepatientsarealworldstudy AT qianmeijiang aclinicalpredictionmodelbasedoninterpretablemachinelearningalgorithmsforprolongedhospitalstayinacuteischemicstrokepatientsarealworldstudy AT muronggao aclinicalpredictionmodelbasedoninterpretablemachinelearningalgorithmsforprolongedhospitalstayinacuteischemicstrokepatientsarealworldstudy AT xiuewei aclinicalpredictionmodelbasedoninterpretablemachinelearningalgorithmsforprolongedhospitalstayinacuteischemicstrokepatientsarealworldstudy AT xiuewei aclinicalpredictionmodelbasedoninterpretablemachinelearningalgorithmsforprolongedhospitalstayinacuteischemicstrokepatientsarealworldstudy AT chanxu aclinicalpredictionmodelbasedoninterpretablemachinelearningalgorithmsforprolongedhospitalstayinacuteischemicstrokepatientsarealworldstudy AT chengliangyin aclinicalpredictionmodelbasedoninterpretablemachinelearningalgorithmsforprolongedhospitalstayinacuteischemicstrokepatientsarealworldstudy AT haiyanliu aclinicalpredictionmodelbasedoninterpretablemachinelearningalgorithmsforprolongedhospitalstayinacuteischemicstrokepatientsarealworldstudy AT haiyanliu aclinicalpredictionmodelbasedoninterpretablemachinelearningalgorithmsforprolongedhospitalstayinacuteischemicstrokepatientsarealworldstudy AT renjungu aclinicalpredictionmodelbasedoninterpretablemachinelearningalgorithmsforprolongedhospitalstayinacuteischemicstrokepatientsarealworldstudy AT haoshengwang aclinicalpredictionmodelbasedoninterpretablemachinelearningalgorithmsforprolongedhospitalstayinacuteischemicstrokepatientsarealworldstudy AT haoshengwang aclinicalpredictionmodelbasedoninterpretablemachinelearningalgorithmsforprolongedhospitalstayinacuteischemicstrokepatientsarealworldstudy AT wenleli aclinicalpredictionmodelbasedoninterpretablemachinelearningalgorithmsforprolongedhospitalstayinacuteischemicstrokepatientsarealworldstudy AT wenleli aclinicalpredictionmodelbasedoninterpretablemachinelearningalgorithmsforprolongedhospitalstayinacuteischemicstrokepatientsarealworldstudy AT liangqunrong aclinicalpredictionmodelbasedoninterpretablemachinelearningalgorithmsforprolongedhospitalstayinacuteischemicstrokepatientsarealworldstudy AT liangqunrong aclinicalpredictionmodelbasedoninterpretablemachinelearningalgorithmsforprolongedhospitalstayinacuteischemicstrokepatientsarealworldstudy AT kaiwang clinicalpredictionmodelbasedoninterpretablemachinelearningalgorithmsforprolongedhospitalstayinacuteischemicstrokepatientsarealworldstudy AT kaiwang clinicalpredictionmodelbasedoninterpretablemachinelearningalgorithmsforprolongedhospitalstayinacuteischemicstrokepatientsarealworldstudy AT qianmeijiang clinicalpredictionmodelbasedoninterpretablemachinelearningalgorithmsforprolongedhospitalstayinacuteischemicstrokepatientsarealworldstudy AT muronggao clinicalpredictionmodelbasedoninterpretablemachinelearningalgorithmsforprolongedhospitalstayinacuteischemicstrokepatientsarealworldstudy AT xiuewei clinicalpredictionmodelbasedoninterpretablemachinelearningalgorithmsforprolongedhospitalstayinacuteischemicstrokepatientsarealworldstudy AT xiuewei clinicalpredictionmodelbasedoninterpretablemachinelearningalgorithmsforprolongedhospitalstayinacuteischemicstrokepatientsarealworldstudy AT chanxu clinicalpredictionmodelbasedoninterpretablemachinelearningalgorithmsforprolongedhospitalstayinacuteischemicstrokepatientsarealworldstudy AT chengliangyin clinicalpredictionmodelbasedoninterpretablemachinelearningalgorithmsforprolongedhospitalstayinacuteischemicstrokepatientsarealworldstudy AT haiyanliu clinicalpredictionmodelbasedoninterpretablemachinelearningalgorithmsforprolongedhospitalstayinacuteischemicstrokepatientsarealworldstudy AT haiyanliu clinicalpredictionmodelbasedoninterpretablemachinelearningalgorithmsforprolongedhospitalstayinacuteischemicstrokepatientsarealworldstudy AT renjungu clinicalpredictionmodelbasedoninterpretablemachinelearningalgorithmsforprolongedhospitalstayinacuteischemicstrokepatientsarealworldstudy AT haoshengwang clinicalpredictionmodelbasedoninterpretablemachinelearningalgorithmsforprolongedhospitalstayinacuteischemicstrokepatientsarealworldstudy AT haoshengwang clinicalpredictionmodelbasedoninterpretablemachinelearningalgorithmsforprolongedhospitalstayinacuteischemicstrokepatientsarealworldstudy AT wenleli clinicalpredictionmodelbasedoninterpretablemachinelearningalgorithmsforprolongedhospitalstayinacuteischemicstrokepatientsarealworldstudy AT wenleli clinicalpredictionmodelbasedoninterpretablemachinelearningalgorithmsforprolongedhospitalstayinacuteischemicstrokepatientsarealworldstudy AT liangqunrong clinicalpredictionmodelbasedoninterpretablemachinelearningalgorithmsforprolongedhospitalstayinacuteischemicstrokepatientsarealworldstudy AT liangqunrong clinicalpredictionmodelbasedoninterpretablemachinelearningalgorithmsforprolongedhospitalstayinacuteischemicstrokepatientsarealworldstudy |