Jin-Zhou Feng1, Yu Wang2, Jin Peng3, Ming-Wei Sun4, Jun Zeng5, Hua Jiang6. 1. Institute for Emergency and Disaster Medicine, Sichuan Academy of Medical Sciences, Sichuan Provincial People's Hospital, School of Medicine, University of Electronic Science and Technology of China, No. 32, Yi Huan Lu Xi Er Duan, Chengdu, Sichuan Province, China; Sino-Finnish Medical AI Research Center, No. 32, Yi Huan Lu Xi Er Duan, Chengdu, Sichuan Province, China. Electronic address: feng720930@126.com. 2. Institute for Emergency and Disaster Medicine, Sichuan Academy of Medical Sciences, Sichuan Provincial People's Hospital, School of Medicine, University of Electronic Science and Technology of China, No. 32, Yi Huan Lu Xi Er Duan, Chengdu, Sichuan Province, China; Sino-Finnish Medical AI Research Center, No. 32, Yi Huan Lu Xi Er Duan, Chengdu, Sichuan Province, China. Electronic address: wangyu@traumabank.org. 3. Institute for Emergency and Disaster Medicine, Sichuan Academy of Medical Sciences, Sichuan Provincial People's Hospital, School of Medicine, University of Electronic Science and Technology of China, No. 32, Yi Huan Lu Xi Er Duan, Chengdu, Sichuan Province, China; Sino-Finnish Medical AI Research Center, No. 32, Yi Huan Lu Xi Er Duan, Chengdu, Sichuan Province, China; Department of Histology, Embryology and Neurobiology, West China School of Basic Medical Sciences & Forensic Medicine, Sichuan University, No. 17, People's South Road, Chengdu, Sichuan, China. Electronic address: admin@traumabank.org. 4. Institute for Emergency and Disaster Medicine, Sichuan Academy of Medical Sciences, Sichuan Provincial People's Hospital, School of Medicine, University of Electronic Science and Technology of China, No. 32, Yi Huan Lu Xi Er Duan, Chengdu, Sichuan Province, China. Electronic address: 653176424@qq.com. 5. Institute for Emergency and Disaster Medicine, Sichuan Academy of Medical Sciences, Sichuan Provincial People's Hospital, School of Medicine, University of Electronic Science and Technology of China, No. 32, Yi Huan Lu Xi Er Duan, Chengdu, Sichuan Province, China; Sino-Finnish Medical AI Research Center, No. 32, Yi Huan Lu Xi Er Duan, Chengdu, Sichuan Province, China. Electronic address: zengjun@medmail.com.cn. 6. Institute for Emergency and Disaster Medicine, Sichuan Academy of Medical Sciences, Sichuan Provincial People's Hospital, School of Medicine, University of Electronic Science and Technology of China, No. 32, Yi Huan Lu Xi Er Duan, Chengdu, Sichuan Province, China; Sino-Finnish Medical AI Research Center, No. 32, Yi Huan Lu Xi Er Duan, Chengdu, Sichuan Province, China. Electronic address: hua.jiang@traumabank.org.
Abstract
PURPOSE: To compare twenty-two machine learning (ML) models against logistic regression on survival prediction in severe traumatic brain injury (STBI) patients in a single center study. MATERIALS AND METHODS: Data was collected from STBI patients admitted to the Sichuan Provincial People's Hospital between December 2009 and November 2011. Twenty-two machine learning (ML) models were tested, and their predictive performance compared with logistic regression (LR) model. Receiver operating characteristics (ROC), area under curve (AUC), accuracy, F-score, precision, recall and Decision Curve Analysis (DCA) were used as performance metrics. RESULTS: A total of 117 patients were enrolled. AUC of all ML models ranged from 86.3% to 94%. AUC of LR was 83%, and accuracy was 88%. The AUC of Cubic SVM, Quadratic SVM and Linear SVM were higher than that of LR. The precision ratio of LR was 95% and recall ratio was 91%, both were lower than most ML models. The F-Score of LR was 0.93, which was only slightly better than that of Linear Discriminant and Quadratic Discriminant. CONCLUSIONS: The twenty-two ML models selected have capabilities comparable to classical LR model for outcome prediction in STBI patients. Of these, Cubic SVM, Quadratic SVM, Linear SVM performed significantly better than LR.
PURPOSE: To compare twenty-two machine learning (ML) models against logistic regression on survival prediction in severe traumatic brain injury (STBI) patients in a single center study. MATERIALS AND METHODS: Data was collected from STBI patients admitted to the Sichuan Provincial People's Hospital between December 2009 and November 2011. Twenty-two machine learning (ML) models were tested, and their predictive performance compared with logistic regression (LR) model. Receiver operating characteristics (ROC), area under curve (AUC), accuracy, F-score, precision, recall and Decision Curve Analysis (DCA) were used as performance metrics. RESULTS: A total of 117 patients were enrolled. AUC of all ML models ranged from 86.3% to 94%. AUC of LR was 83%, and accuracy was 88%. The AUC of Cubic SVM, Quadratic SVM and Linear SVM were higher than that of LR. The precision ratio of LR was 95% and recall ratio was 91%, both were lower than most ML models. The F-Score of LR was 0.93, which was only slightly better than that of Linear Discriminant and Quadratic Discriminant. CONCLUSIONS: The twenty-two ML models selected have capabilities comparable to classical LR model for outcome prediction in STBI patients. Of these, Cubic SVM, Quadratic SVM, Linear SVM performed significantly better than LR.
Authors: Ahmed Abdulaal; Aatish Patel; Esmita Charani; Sarah Denny; Saleh A Alqahtani; Gary W Davies; Nabeela Mughal; Luke S P Moore Journal: BMC Med Inform Decis Mak Date: 2020-11-19 Impact factor: 2.796
Authors: Jose M Celaya-Padilla; Karen E Villagrana-Bañuelos; Juan José Oropeza-Valdez; Joel Monárrez-Espino; Julio E Castañeda-Delgado; Ana Sofía Herrera-Van Oostdam; Julio César Fernández-Ruiz; Fátima Ochoa-González; Juan Carlos Borrego; Jose Antonio Enciso-Moreno; Jesús Adrián López; Yamilé López-Hernández; Carlos E Galván-Tejada Journal: Diagnostics (Basel) Date: 2021-11-25