Yue Gao1,2,3, Lingxi Chen4, Jianhua Chi1,2,3, Shaoqing Zeng1,2,3, Xikang Feng5, Huayi Li1,2,3, Dan Liu1,2,3, Xinxia Feng6, Siyuan Wang1,2,3, Ya Wang1,2,3, Ruidi Yu1,2,3, Yuan Yuan1,2,3, Sen Xu1,2,3, Chunrui Li7, Wei Zhang8,9,10, Shuaicheng Li11, Qinglei Gao12,13,14. 1. National Medical Center for Major Public Health Events, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, 430000, China. 2. Cancer Biology Research Center (Key Laboratory of Chinese Ministry of Education), Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, 1095 Jiefang Ave, Wuhan, 430000, China. 3. Department of Gynecology and Obstetrics, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, People's Republic of China. 4. Department of Computer Science, City University of Hong Kong, Tatchee Avenue, Kowloon Tong, 999077, Hong Kong. 5. School of Software, Northwestern Polytechnical University, Xi'an, People's Republic of China. 6. Department of Gastroenterology, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, 430000, People's Republic of China. 7. Department of Hematology, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, People's Republic of China. 8. National Medical Center for Major Public Health Events, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, 430000, China. zhangwei_tjh@qq.com. 9. Cancer Biology Research Center (Key Laboratory of Chinese Ministry of Education), Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, 1095 Jiefang Ave, Wuhan, 430000, China. zhangwei_tjh@qq.com. 10. Department of Gynecology and Obstetrics, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, People's Republic of China. zhangwei_tjh@qq.com. 11. Department of Computer Science, City University of Hong Kong, Tatchee Avenue, Kowloon Tong, 999077, Hong Kong. shuaicli@cityu.edu.hk. 12. National Medical Center for Major Public Health Events, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, 430000, China. qingleigao@hotmail.com. 13. Cancer Biology Research Center (Key Laboratory of Chinese Ministry of Education), Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, 1095 Jiefang Ave, Wuhan, 430000, China. qingleigao@hotmail.com. 14. Department of Gynecology and Obstetrics, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, People's Republic of China. qingleigao@hotmail.com.
Abstract
BACKGROUND: Immune and inflammatory dysfunction was reported to underpin critical COVID-19(coronavirus disease 2019). We aim to develop a machine learning model that enables accurate prediction of critical COVID-19 using immune-inflammatory features at admission. METHODS: We retrospectively collected 2076 consecutive COVID-19 patients with definite outcomes (discharge or death) between January 27, 2020 and March 30, 2020 from two hospitals in China. Critical illness was defined as admission to intensive care unit, receiving invasive ventilation, or death. Least Absolute Shrinkage and Selection Operator (LASSO) was applied for feature selection. Five machine learning algorithms, including Logistic Regression (LR), Support Vector Machine (SVM), Gradient Boosted Decision Tree (GBDT), K-Nearest Neighbor (KNN), and Neural Network (NN) were built in a training dataset, and assessed in an internal validation dataset and an external validation dataset. RESULTS: Six features (procalcitonin, [T + B + NK cell] count, interleukin 6, C reactive protein, interleukin 2 receptor, T-helper lymphocyte/T-suppressor lymphocyte) were finally used for model development. Five models displayed varying but all promising predictive performance. Notably, the ensemble model, SPMCIIP (severity prediction model for COVID-19 by immune-inflammatory parameters), derived from three contributive algorithms (SVM, GBDT, and NN) achieved the best performance with an area under the curve (AUC) of 0.991 (95% confidence interval [CI] 0.979-1.000) in internal validation cohort and 0.999 (95% CI 0.998-1.000) in external validation cohort to identify patients with critical COVID-19. SPMCIIP could accurately and expeditiously predict the occurrence of critical COVID-19 approximately 20 days in advance. CONCLUSIONS: The developed online prediction model SPMCIIP is hopeful to facilitate intensive monitoring and early intervention of high risk of critical illness in COVID-19 patients. TRIAL REGISTRATION: This study was retrospectively registered in the Chinese Clinical Trial Registry ( ChiCTR2000032161 ). vv.
BACKGROUND: Immune and inflammatory dysfunction was reported to underpin critical COVID-19(coronavirus disease 2019). We aim to develop a machine learning model that enables accurate prediction of critical COVID-19 using immune-inflammatory features at admission. METHODS: We retrospectively collected 2076 consecutive COVID-19patients with definite outcomes (discharge or death) between January 27, 2020 and March 30, 2020 from two hospitals in China. Critical illness was defined as admission to intensive care unit, receiving invasive ventilation, or death. Least Absolute Shrinkage and Selection Operator (LASSO) was applied for feature selection. Five machine learning algorithms, including Logistic Regression (LR), Support Vector Machine (SVM), Gradient Boosted Decision Tree (GBDT), K-Nearest Neighbor (KNN), and Neural Network (NN) were built in a training dataset, and assessed in an internal validation dataset and an external validation dataset. RESULTS: Six features (procalcitonin, [T + B + NK cell] count, interleukin 6, C reactive protein, interleukin 2 receptor, T-helper lymphocyte/T-suppressor lymphocyte) were finally used for model development. Five models displayed varying but all promising predictive performance. Notably, the ensemble model, SPMCIIP (severity prediction model for COVID-19 by immune-inflammatory parameters), derived from three contributive algorithms (SVM, GBDT, and NN) achieved the best performance with an area under the curve (AUC) of 0.991 (95% confidence interval [CI] 0.979-1.000) in internal validation cohort and 0.999 (95% CI 0.998-1.000) in external validation cohort to identify patients with critical COVID-19. SPMCIIP could accurately and expeditiously predict the occurrence of critical COVID-19 approximately 20 days in advance. CONCLUSIONS: The developed online prediction model SPMCIIP is hopeful to facilitate intensive monitoring and early intervention of high risk of critical illness in COVID-19patients. TRIAL REGISTRATION: This study was retrospectively registered in the Chinese Clinical Trial Registry ( ChiCTR2000032161 ). vv.
Entities:
Keywords:
COVID-19; Critical illness; Immune-inflammatory parameters; Machine learning; Online model
Authors: Pauline Heus; Johannes B Reitsma; Gary S Collins; Johanna A A G Damen; Rob J P M Scholten; Douglas G Altman; Karel G M Moons; Lotty Hooft Journal: Ann Intern Med Date: 2020-06-02 Impact factor: 25.391
Authors: Leticia Kuri-Cervantes; M Betina Pampena; Wenzhao Meng; Aaron M Rosenfeld; Caroline A G Ittner; Ariel R Weisman; Roseline S Agyekum; Divij Mathew; Amy E Baxter; Laura A Vella; Oliva Kuthuru; Sokratis A Apostolidis; Luanne Bershaw; Jeanette Dougherty; Allison R Greenplate; Ajinkya Pattekar; Justin Kim; Nicholas Han; Sigrid Gouma; Madison E Weirick; Claudia P Arevalo; Marcus J Bolton; Eileen C Goodwin; Elizabeth M Anderson; Scott E Hensley; Tiffanie K Jones; Nilam S Mangalmurti; Eline T Luning Prak; E John Wherry; Nuala J Meyer; Michael R Betts Journal: Sci Immunol Date: 2020-07-15
Authors: Puja Mehta; Daniel F McAuley; Michael Brown; Emilie Sanchez; Rachel S Tattersall; Jessica J Manson Journal: Lancet Date: 2020-03-16 Impact factor: 79.321