Supervised machine learning algorithms to predict the duration and risk of long-term hospitalization in HIV-infected individuals: a retrospective study
ObjectiveThe study aimed to use supervised machine learning models to predict the length and risk of prolonged hospitalization in PLWHs to help physicians timely clinical intervention and avoid waste of health resources.MethodsRegression models were established based on RF, KNN, SVM, and XGB to pred...
| Main Authors: | , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2024-01-01
|
| Series: | Frontiers in Public Health |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fpubh.2023.1282324/full |
| _version_ | 1797365520194863104 |
|---|---|
| author | Jialu Li Yiwei Hao Ying Liu Liang Wu Hongyuan Liang Liang Ni Fang Wang Sa Wang Yujiao Duan Qiuhua Xu Jinjing Xiao Di Yang Guiju Gao Yi Ding Chengyu Gao Jiang Xiao Hongxin Zhao |
| author_facet | Jialu Li Yiwei Hao Ying Liu Liang Wu Hongyuan Liang Liang Ni Fang Wang Sa Wang Yujiao Duan Qiuhua Xu Jinjing Xiao Di Yang Guiju Gao Yi Ding Chengyu Gao Jiang Xiao Hongxin Zhao |
| author_sort | Jialu Li |
| collection | DOAJ |
| description | ObjectiveThe study aimed to use supervised machine learning models to predict the length and risk of prolonged hospitalization in PLWHs to help physicians timely clinical intervention and avoid waste of health resources.MethodsRegression models were established based on RF, KNN, SVM, and XGB to predict the length of hospital stay using RMSE, MAE, MAPE, and R2, while classification models were established based on RF, KNN, SVM, NN, and XGB to predict risk of prolonged hospital stay using accuracy, PPV, NPV, specificity, sensitivity, and kappa, and visualization evaluation based on AUROC, AUPRC, calibration curves and decision curves of all models were used for internally validation.ResultsIn regression models, XGB model performed best in the internal validation (RMSE = 16.81, MAE = 10.39, MAPE = 0.98, R2 = 0.47) to predict the length of hospital stay, while in classification models, NN model presented good fitting and stable features and performed best in testing sets, with excellent accuracy (0.7623), PPV (0.7853), NPV (0.7092), sensitivity (0.8754), specificity (0.5882), and kappa (0.4672), and further visualization evaluation indicated that the largest AUROC (0.9779), AUPRC (0.773) and well-performed calibration curve and decision curve in the internal validation.ConclusionThis study showed that XGB model was effective in predicting the length of hospital stay, while NN model was effective in predicting the risk of prolonged hospitalization in PLWH. Based on predictive models, an intelligent medical prediction system may be developed to effectively predict the length of stay and risk of HIV patients according to their medical records, which helped reduce the waste of healthcare resources. |
| first_indexed | 2024-03-08T16:50:10Z |
| format | Article |
| id | doaj.art-8f39b33fa5fb4e81823820a8f1e159b9 |
| institution | Directory Open Access Journal |
| issn | 2296-2565 |
| language | English |
| last_indexed | 2024-03-08T16:50:10Z |
| publishDate | 2024-01-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Public Health |
| spelling | doaj.art-8f39b33fa5fb4e81823820a8f1e159b92024-01-05T04:57:10ZengFrontiers Media S.A.Frontiers in Public Health2296-25652024-01-011110.3389/fpubh.2023.12823241282324Supervised machine learning algorithms to predict the duration and risk of long-term hospitalization in HIV-infected individuals: a retrospective studyJialu Li0Yiwei Hao1Ying Liu2Liang Wu3Hongyuan Liang4Liang Ni5Fang Wang6Sa Wang7Yujiao Duan8Qiuhua Xu9Jinjing Xiao10Di Yang11Guiju Gao12Yi Ding13Chengyu Gao14Jiang Xiao15Hongxin Zhao16Clinical and Research Center of AIDS, Beijing Ditan Hospital, Capital Medical University, Beijing, ChinaDivision of Medical Record and Statistics, Beijing Ditan Hospital, Capital Medical University, Beijing, ChinaClinical and Research Center of AIDS, Beijing Ditan Hospital, Capital Medical University, Beijing, ChinaClinical and Research Center of AIDS, Beijing Ditan Hospital, Capital Medical University, Beijing, ChinaClinical and Research Center of AIDS, Beijing Ditan Hospital, Capital Medical University, Beijing, ChinaClinical and Research Center of AIDS, Beijing Ditan Hospital, Capital Medical University, Beijing, ChinaClinical and Research Center of AIDS, Beijing Ditan Hospital, Capital Medical University, Beijing, ChinaClinical and Research Center of AIDS, Beijing Ditan Hospital, Capital Medical University, Beijing, ChinaClinical and Research Center of AIDS, Beijing Ditan Hospital, Capital Medical University, Beijing, ChinaClinical and Research Center of AIDS, Beijing Ditan Hospital, Capital Medical University, Beijing, ChinaDepartment of Clinical Medicine, Zhengzhou University, Zhengzhou, ChinaClinical and Research Center of AIDS, Beijing Ditan Hospital, Capital Medical University, Beijing, ChinaClinical and Research Center of AIDS, Beijing Ditan Hospital, Capital Medical University, Beijing, ChinaClinical and Research Center of AIDS, Beijing Ditan Hospital, Capital Medical University, Beijing, ChinaClinical and Research Center of AIDS, Beijing Ditan Hospital, Capital Medical University, Beijing, ChinaClinical and Research Center of AIDS, Beijing Ditan Hospital, Capital Medical University, Beijing, ChinaClinical and Research Center of AIDS, Beijing Ditan Hospital, Capital Medical University, Beijing, ChinaObjectiveThe study aimed to use supervised machine learning models to predict the length and risk of prolonged hospitalization in PLWHs to help physicians timely clinical intervention and avoid waste of health resources.MethodsRegression models were established based on RF, KNN, SVM, and XGB to predict the length of hospital stay using RMSE, MAE, MAPE, and R2, while classification models were established based on RF, KNN, SVM, NN, and XGB to predict risk of prolonged hospital stay using accuracy, PPV, NPV, specificity, sensitivity, and kappa, and visualization evaluation based on AUROC, AUPRC, calibration curves and decision curves of all models were used for internally validation.ResultsIn regression models, XGB model performed best in the internal validation (RMSE = 16.81, MAE = 10.39, MAPE = 0.98, R2 = 0.47) to predict the length of hospital stay, while in classification models, NN model presented good fitting and stable features and performed best in testing sets, with excellent accuracy (0.7623), PPV (0.7853), NPV (0.7092), sensitivity (0.8754), specificity (0.5882), and kappa (0.4672), and further visualization evaluation indicated that the largest AUROC (0.9779), AUPRC (0.773) and well-performed calibration curve and decision curve in the internal validation.ConclusionThis study showed that XGB model was effective in predicting the length of hospital stay, while NN model was effective in predicting the risk of prolonged hospitalization in PLWH. Based on predictive models, an intelligent medical prediction system may be developed to effectively predict the length of stay and risk of HIV patients according to their medical records, which helped reduce the waste of healthcare resources.https://www.frontiersin.org/articles/10.3389/fpubh.2023.1282324/fullHIVAIDSmachine learninglength of stayrisk factorscalibration curves |
| spellingShingle | Jialu Li Yiwei Hao Ying Liu Liang Wu Hongyuan Liang Liang Ni Fang Wang Sa Wang Yujiao Duan Qiuhua Xu Jinjing Xiao Di Yang Guiju Gao Yi Ding Chengyu Gao Jiang Xiao Hongxin Zhao Supervised machine learning algorithms to predict the duration and risk of long-term hospitalization in HIV-infected individuals: a retrospective study Frontiers in Public Health HIV AIDS machine learning length of stay risk factors calibration curves |
| title | Supervised machine learning algorithms to predict the duration and risk of long-term hospitalization in HIV-infected individuals: a retrospective study |
| title_full | Supervised machine learning algorithms to predict the duration and risk of long-term hospitalization in HIV-infected individuals: a retrospective study |
| title_fullStr | Supervised machine learning algorithms to predict the duration and risk of long-term hospitalization in HIV-infected individuals: a retrospective study |
| title_full_unstemmed | Supervised machine learning algorithms to predict the duration and risk of long-term hospitalization in HIV-infected individuals: a retrospective study |
| title_short | Supervised machine learning algorithms to predict the duration and risk of long-term hospitalization in HIV-infected individuals: a retrospective study |
| title_sort | supervised machine learning algorithms to predict the duration and risk of long term hospitalization in hiv infected individuals a retrospective study |
| topic | HIV AIDS machine learning length of stay risk factors calibration curves |
| url | https://www.frontiersin.org/articles/10.3389/fpubh.2023.1282324/full |
| work_keys_str_mv | AT jialuli supervisedmachinelearningalgorithmstopredictthedurationandriskoflongtermhospitalizationinhivinfectedindividualsaretrospectivestudy AT yiweihao supervisedmachinelearningalgorithmstopredictthedurationandriskoflongtermhospitalizationinhivinfectedindividualsaretrospectivestudy AT yingliu supervisedmachinelearningalgorithmstopredictthedurationandriskoflongtermhospitalizationinhivinfectedindividualsaretrospectivestudy AT liangwu supervisedmachinelearningalgorithmstopredictthedurationandriskoflongtermhospitalizationinhivinfectedindividualsaretrospectivestudy AT hongyuanliang supervisedmachinelearningalgorithmstopredictthedurationandriskoflongtermhospitalizationinhivinfectedindividualsaretrospectivestudy AT liangni supervisedmachinelearningalgorithmstopredictthedurationandriskoflongtermhospitalizationinhivinfectedindividualsaretrospectivestudy AT fangwang supervisedmachinelearningalgorithmstopredictthedurationandriskoflongtermhospitalizationinhivinfectedindividualsaretrospectivestudy AT sawang supervisedmachinelearningalgorithmstopredictthedurationandriskoflongtermhospitalizationinhivinfectedindividualsaretrospectivestudy AT yujiaoduan supervisedmachinelearningalgorithmstopredictthedurationandriskoflongtermhospitalizationinhivinfectedindividualsaretrospectivestudy AT qiuhuaxu supervisedmachinelearningalgorithmstopredictthedurationandriskoflongtermhospitalizationinhivinfectedindividualsaretrospectivestudy AT jinjingxiao supervisedmachinelearningalgorithmstopredictthedurationandriskoflongtermhospitalizationinhivinfectedindividualsaretrospectivestudy AT diyang supervisedmachinelearningalgorithmstopredictthedurationandriskoflongtermhospitalizationinhivinfectedindividualsaretrospectivestudy AT guijugao supervisedmachinelearningalgorithmstopredictthedurationandriskoflongtermhospitalizationinhivinfectedindividualsaretrospectivestudy AT yiding supervisedmachinelearningalgorithmstopredictthedurationandriskoflongtermhospitalizationinhivinfectedindividualsaretrospectivestudy AT chengyugao supervisedmachinelearningalgorithmstopredictthedurationandriskoflongtermhospitalizationinhivinfectedindividualsaretrospectivestudy AT jiangxiao supervisedmachinelearningalgorithmstopredictthedurationandriskoflongtermhospitalizationinhivinfectedindividualsaretrospectivestudy AT hongxinzhao supervisedmachinelearningalgorithmstopredictthedurationandriskoflongtermhospitalizationinhivinfectedindividualsaretrospectivestudy |