TY - JOUR
T1 - On Classifying Sepsis Heterogeneity in the ICU: Insight Using Machine Learning
AU - Ibrahim, Zina
AU - Wu, Honghan
AU - Hamoud, Ahmed
AU - Stappen, Lukas
AU - Dobson, Richard James Butler
AU - Agarossi, Andrea
PY - 2019/11/24
Y1 - 2019/11/24
N2 - Current machine learning models aiming to predict sepsis from Electronic Health Records (EHR) do not account for the heterogeneity of the condition, despite its emerging importance in prognosis and treatment. This work demonstrates the added value of stratifying the types of organ dysfunction observed in patients who develop sepsis in the ICU in improving the ability to recognise patients at risk of sepsis from their EHR data. Using an ICU dataset of 13,728 records, we identify clinically significant sepsis subpopulations with distinct organ dysfunction patterns. Classification experiments using Random Forest, Gradient Boost Trees and Support Vector Machines, aiming to distinguish patients who develop sepsis in the ICU from those who do not, show that features selected using sepsis subpopulations as background knowledge yield a superior performance regardless of the classification model used. Our findings can steer machine learning efforts towards more personalised models for complex conditions including sepsis.
AB - Current machine learning models aiming to predict sepsis from Electronic Health Records (EHR) do not account for the heterogeneity of the condition, despite its emerging importance in prognosis and treatment. This work demonstrates the added value of stratifying the types of organ dysfunction observed in patients who develop sepsis in the ICU in improving the ability to recognise patients at risk of sepsis from their EHR data. Using an ICU dataset of 13,728 records, we identify clinically significant sepsis subpopulations with distinct organ dysfunction patterns. Classification experiments using Random Forest, Gradient Boost Trees and Support Vector Machines, aiming to distinguish patients who develop sepsis in the ICU from those who do not, show that features selected using sepsis subpopulations as background knowledge yield a superior performance regardless of the classification model used. Our findings can steer machine learning efforts towards more personalised models for complex conditions including sepsis.
U2 - 10.1093/jamia/ocz211
DO - 10.1093/jamia/ocz211
M3 - Article
SN - 1527-974X
JO - Journal of the American Medical Informatics Association : JAMIA
JF - Journal of the American Medical Informatics Association : JAMIA
ER -