Franco van Wyk1, Anahita Khojandi1, Akram Mohammed2, Edmon Begoli3, Robert L Davis2, Rishikesan Kamaleswaran4. 1. University of Tennessee, Knoxville, TN, USA. 2. Center for Biomedical Informatics, Department of Pediatrics, University of Tennessee Health, USA Science Center, Memphis, TN, USA. 3. University of Tennessee, Knoxville, TN, USA; Oak Ridge National Laboratory, Knoxville, TN, USA. 4. Center for Biomedical Informatics, Department of Pediatrics, University of Tennessee Health, USA Science Center, Memphis, TN, USA. Electronic address: rkamales@uthsc.edu.
Abstract
PURPOSE: Sepsis is a life-threatening condition with high mortality rates and expensive treatment costs. To improve short- and long-term outcomes, it is critical to detect at-risk sepsis patients at an early stage. METHODS: A data-set consisting of high-frequency physiological data from 1161 critically ill patients was analyzed. 377 patients had developed sepsis, and had data at least 3 h prior to the onset of sepsis. A random forest classifier was trained to discriminate between sepsis and non-sepsis patients in real-time using a total of 132 features extracted from a moving time-window. The model was trained on 80% of the patients and was tested on the remaining 20% of the patients, for two observational periods of lengths 3 and 6 h prior to onset. RESULTS: The model that used continuous physiological data alone resulted in sensitivity and F1 score of up to 80% and 67% one hour before sepsis onset. On average, these models were able to predict sepsis 294.19 ± 6.50 min (5 h) before the onset. CONCLUSIONS: The use of machine learning algorithms on continuous streams of physiological data can allow for early identification of at-risk patients in real-time with high accuracy.
PURPOSE:Sepsis is a life-threatening condition with high mortality rates and expensive treatment costs. To improve short- and long-term outcomes, it is critical to detect at-risk sepsispatients at an early stage. METHODS: A data-set consisting of high-frequency physiological data from 1161 critically ill patients was analyzed. 377 patients had developed sepsis, and had data at least 3 h prior to the onset of sepsis. A random forest classifier was trained to discriminate between sepsis and non-sepsispatients in real-time using a total of 132 features extracted from a moving time-window. The model was trained on 80% of the patients and was tested on the remaining 20% of the patients, for two observational periods of lengths 3 and 6 h prior to onset. RESULTS: The model that used continuous physiological data alone resulted in sensitivity and F1 score of up to 80% and 67% one hour before sepsis onset. On average, these models were able to predict sepsis 294.19 ± 6.50 min (5 h) before the onset. CONCLUSIONS: The use of machine learning algorithms on continuous streams of physiological data can allow for early identification of at-risk patients in real-time with high accuracy.
Authors: Akram Mohammed; Pradeep S B Podila; Robert L Davis; Kenneth I Ataga; Jane S Hankins; Rishikesan Kamaleswaran Journal: J Med Internet Res Date: 2020-05-13 Impact factor: 5.428
Authors: Daniele Roberto Giacobbe; Alessio Signori; Filippo Del Puente; Sara Mora; Luca Carmisciano; Federica Briano; Antonio Vena; Lorenzo Ball; Chiara Robba; Paolo Pelosi; Mauro Giacomini; Matteo Bassetti Journal: Front Med (Lausanne) Date: 2021-02-12
Authors: Lucas M Fleuren; Thomas L T Klausch; Charlotte L Zwager; Linda J Schoonmade; Tingjie Guo; Luca F Roggeveen; Eleonora L Swart; Armand R J Girbes; Patrick Thoral; Ari Ercole; Mark Hoogendoorn; Paul W G Elbers Journal: Intensive Care Med Date: 2020-01-21 Impact factor: 17.440
Authors: Asher A Mendelson; Ajay Rajaram; Daniel Bainbridge; Keith St Lawrence; Tracey Bentall; Michael Sharpe; Mamadou Diop; Christopher G Ellis Journal: J Clin Monit Comput Date: 2020-10-26 Impact factor: 1.977