Multidimensional dynamic prediction model for hospitalized patients with the omicron variant in China
Purpose: To establish dynamic prediction models by machine learning using daily multidimensional data for coronavirus disease 2019 (COVID-19) patients. Methods: Hospitalized COVID-19 patients at Peking Union Medical College Hospital from Nov 2nd, 2022, to Jan 13th, 2023, were enrolled in this study....
| Main Authors: | , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
KeAi Communications Co., Ltd.
2023-12-01
|
| Series: | Infectious Disease Modelling |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2468042723000854 |
| _version_ | 1797448063207342080 |
|---|---|
| author | Yujie Chen Yao Wang Jieqing Chen Xudong Ma Longxiang Su Yuna Wei Linfeng Li Dandan Ma Feng Zhang Wen Zhu Xiaoyang Meng Guoqiang Sun Lian Ma Huizhen Jiang Chang Yin Taisheng Li Xiang Zhou |
| author_facet | Yujie Chen Yao Wang Jieqing Chen Xudong Ma Longxiang Su Yuna Wei Linfeng Li Dandan Ma Feng Zhang Wen Zhu Xiaoyang Meng Guoqiang Sun Lian Ma Huizhen Jiang Chang Yin Taisheng Li Xiang Zhou |
| author_sort | Yujie Chen |
| collection | DOAJ |
| description | Purpose: To establish dynamic prediction models by machine learning using daily multidimensional data for coronavirus disease 2019 (COVID-19) patients. Methods: Hospitalized COVID-19 patients at Peking Union Medical College Hospital from Nov 2nd, 2022, to Jan 13th, 2023, were enrolled in this study. The outcome was defined as deterioration or recovery of the patient's condition. Demographics, comorbidities, laboratory test results, vital signs, and treatments were used to train the model. To predict the following days, a separate XGBoost model was trained and validated. The Shapley additive explanations method was used to analyze feature importance. Results: A total of 995 patients were enrolled, generating 7228 and 3170 observations for each prediction model. In the deterioration prediction model, the minimum area under the receiver operating characteristic curve (AUROC) for the following 7 days was 0.786 (95% CI 0.721–0.851), while the AUROC on the next day was 0.872 (0.831–0.913). In the recovery prediction model, the minimum AUROC for the following 3 days was 0.675 (0.583–0.767), while the AUROC on the next day was 0.823 (0.770–0.876). The top 5 features for deterioration prediction on the 7th day were disease course, length of hospital stay, hypertension, and diastolic blood pressure. Those for recovery prediction on the 3rd day were age, D-dimer levels, disease course, creatinine levels and corticosteroid therapy. Conclusion: The models could accurately predict the dynamics of Omicron patients’ conditions using daily multidimensional variables, revealing important features including comorbidities (e.g., hyperlipidemia), age, disease course, vital signs, D-dimer levels, corticosteroid therapy and oxygen therapy. |
| first_indexed | 2024-03-09T14:04:57Z |
| format | Article |
| id | doaj.art-83bb1e9229d249e2b00158ed2c54c9a3 |
| institution | Directory Open Access Journal |
| issn | 2468-0427 |
| language | English |
| last_indexed | 2024-03-09T14:04:57Z |
| publishDate | 2023-12-01 |
| publisher | KeAi Communications Co., Ltd. |
| record_format | Article |
| series | Infectious Disease Modelling |
| spelling | doaj.art-83bb1e9229d249e2b00158ed2c54c9a32023-11-30T05:08:12ZengKeAi Communications Co., Ltd.Infectious Disease Modelling2468-04272023-12-018410971107Multidimensional dynamic prediction model for hospitalized patients with the omicron variant in ChinaYujie Chen0Yao Wang1Jieqing Chen2Xudong Ma3Longxiang Su4Yuna Wei5Linfeng Li6Dandan Ma7Feng Zhang8Wen Zhu9Xiaoyang Meng10Guoqiang Sun11Lian Ma12Huizhen Jiang13Chang Yin14Taisheng Li15Xiang Zhou16Department of Critical Care Medicine, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, 100730, ChinaYidu Cloud Technology Inc., Beijing, ChinaInformation Center Department, State Key Laboratory of Complex Severe and Rare Diseases, Peking Union Medical College Hospital, Peking Union Medical College and Chinese Academy of Medical Sciences, Beijing, 100730, ChinaDepartment of Medical Administration, National Health Commission of the People's Republic of China, Beijing, 100044, ChinaDepartment of Critical Care Medicine, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, 100730, ChinaYidu Cloud Technology Inc., Beijing, ChinaYidu Cloud Technology Inc., Beijing, ChinaInformation Center, State Key Laboratory of Complex Severe and Rare Diseases, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, 100730, ChinaInformation Center, State Key Laboratory of Complex Severe and Rare Diseases, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, 100730, ChinaInformation Center, State Key Laboratory of Complex Severe and Rare Diseases, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, 100730, ChinaInformation Center, State Key Laboratory of Complex Severe and Rare Diseases, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, 100730, ChinaInformation Center, State Key Laboratory of Complex Severe and Rare Diseases, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, 100730, ChinaInformation Center, State Key Laboratory of Complex Severe and Rare Diseases, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, 100730, ChinaInformation Center, State Key Laboratory of Complex Severe and Rare Diseases, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, 100730, ChinaNational Institute of Hospital Administration, Beijing, China; Corresponding author.Department of Infectious Diseases, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences, Beijing, China; Corresponding author.Department of Critical Care Medicine, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, 100730, China; Information Center Department, State Key Laboratory of Complex Severe and Rare Diseases, Peking Union Medical College Hospital, Peking Union Medical College and Chinese Academy of Medical Sciences, Beijing, 100730, China; Corresponding author. Department of Critical Care Medicine, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing,100730, China.Purpose: To establish dynamic prediction models by machine learning using daily multidimensional data for coronavirus disease 2019 (COVID-19) patients. Methods: Hospitalized COVID-19 patients at Peking Union Medical College Hospital from Nov 2nd, 2022, to Jan 13th, 2023, were enrolled in this study. The outcome was defined as deterioration or recovery of the patient's condition. Demographics, comorbidities, laboratory test results, vital signs, and treatments were used to train the model. To predict the following days, a separate XGBoost model was trained and validated. The Shapley additive explanations method was used to analyze feature importance. Results: A total of 995 patients were enrolled, generating 7228 and 3170 observations for each prediction model. In the deterioration prediction model, the minimum area under the receiver operating characteristic curve (AUROC) for the following 7 days was 0.786 (95% CI 0.721–0.851), while the AUROC on the next day was 0.872 (0.831–0.913). In the recovery prediction model, the minimum AUROC for the following 3 days was 0.675 (0.583–0.767), while the AUROC on the next day was 0.823 (0.770–0.876). The top 5 features for deterioration prediction on the 7th day were disease course, length of hospital stay, hypertension, and diastolic blood pressure. Those for recovery prediction on the 3rd day were age, D-dimer levels, disease course, creatinine levels and corticosteroid therapy. Conclusion: The models could accurately predict the dynamics of Omicron patients’ conditions using daily multidimensional variables, revealing important features including comorbidities (e.g., hyperlipidemia), age, disease course, vital signs, D-dimer levels, corticosteroid therapy and oxygen therapy.http://www.sciencedirect.com/science/article/pii/S2468042723000854COVID-19OmicronPrediction modelMachine learning |
| spellingShingle | Yujie Chen Yao Wang Jieqing Chen Xudong Ma Longxiang Su Yuna Wei Linfeng Li Dandan Ma Feng Zhang Wen Zhu Xiaoyang Meng Guoqiang Sun Lian Ma Huizhen Jiang Chang Yin Taisheng Li Xiang Zhou Multidimensional dynamic prediction model for hospitalized patients with the omicron variant in China Infectious Disease Modelling COVID-19 Omicron Prediction model Machine learning |
| title | Multidimensional dynamic prediction model for hospitalized patients with the omicron variant in China |
| title_full | Multidimensional dynamic prediction model for hospitalized patients with the omicron variant in China |
| title_fullStr | Multidimensional dynamic prediction model for hospitalized patients with the omicron variant in China |
| title_full_unstemmed | Multidimensional dynamic prediction model for hospitalized patients with the omicron variant in China |
| title_short | Multidimensional dynamic prediction model for hospitalized patients with the omicron variant in China |
| title_sort | multidimensional dynamic prediction model for hospitalized patients with the omicron variant in china |
| topic | COVID-19 Omicron Prediction model Machine learning |
| url | http://www.sciencedirect.com/science/article/pii/S2468042723000854 |
| work_keys_str_mv | AT yujiechen multidimensionaldynamicpredictionmodelforhospitalizedpatientswiththeomicronvariantinchina AT yaowang multidimensionaldynamicpredictionmodelforhospitalizedpatientswiththeomicronvariantinchina AT jieqingchen multidimensionaldynamicpredictionmodelforhospitalizedpatientswiththeomicronvariantinchina AT xudongma multidimensionaldynamicpredictionmodelforhospitalizedpatientswiththeomicronvariantinchina AT longxiangsu multidimensionaldynamicpredictionmodelforhospitalizedpatientswiththeomicronvariantinchina AT yunawei multidimensionaldynamicpredictionmodelforhospitalizedpatientswiththeomicronvariantinchina AT linfengli multidimensionaldynamicpredictionmodelforhospitalizedpatientswiththeomicronvariantinchina AT dandanma multidimensionaldynamicpredictionmodelforhospitalizedpatientswiththeomicronvariantinchina AT fengzhang multidimensionaldynamicpredictionmodelforhospitalizedpatientswiththeomicronvariantinchina AT wenzhu multidimensionaldynamicpredictionmodelforhospitalizedpatientswiththeomicronvariantinchina AT xiaoyangmeng multidimensionaldynamicpredictionmodelforhospitalizedpatientswiththeomicronvariantinchina AT guoqiangsun multidimensionaldynamicpredictionmodelforhospitalizedpatientswiththeomicronvariantinchina AT lianma multidimensionaldynamicpredictionmodelforhospitalizedpatientswiththeomicronvariantinchina AT huizhenjiang multidimensionaldynamicpredictionmodelforhospitalizedpatientswiththeomicronvariantinchina AT changyin multidimensionaldynamicpredictionmodelforhospitalizedpatientswiththeomicronvariantinchina AT taishengli multidimensionaldynamicpredictionmodelforhospitalizedpatientswiththeomicronvariantinchina AT xiangzhou multidimensionaldynamicpredictionmodelforhospitalizedpatientswiththeomicronvariantinchina |