An Explainable Artificial Intelligence Predictor for Early Detection of Sepsis

Abstract

Objectives: Early detection of sepsis is critical in clinical practice since each hour of delayed treatment has been associated with an increase in mortality due to irreversible organ damage. This study aimed to develop an explainable artificial intelligence model for early predicting sepsis by analyzing the electronic health record data from ICU provided by the PhysioNet/Computing in Cardiology Challenge 2019. Design: Retrospective observational study. Setting: We developed our model on the shared ICUs publicly data and verified on the full hidden populations for challenge scoring. Patients: Public database included 40,336 patients’ electronic health records sourced from Beth Israel Deaconess Medical Center (hospital system A) and Emory University Hospital (hospital system B). A total of 24,819 patients from hospital systems A, B, and C (an unidentified hospital system) were sequestered as full hidden test sets. Interventions: None. Measurements and Main Results: A total of 168 features were extracted on hourly basis. Explainable artificial intelligence sepsis predictor model was trained to predict sepsis in real time. Impact of each feature on hourly sepsis prediction was explored in-depth to show the interpretability. The algorithm demonstrated the final clinical utility score of 0.364 in this challenge when tested on the full hidden test sets, and the scores on three separate test sets were 0.430, 0.422, and –0.048, respectively. Conclusions: Explainable artificial intelligence sepsis predictor model achieves superior performance for predicting sepsis risk in a real-time way and provides interpretable information for understanding sepsis risk in ICU.

Early Prediction of Sepsis Onset Using Neural Architecture Search Based on Genetic Algorithms

“…These data are sent through a noise removal process using a trained model, followed by a prediction process. Table 4 shows the sepsis onset prediction results which were obtained by the various models, including rule-based scoring systems [ 14 , 20 , 24 , 30 , 31 , 32 , 33 , 34 , 35 , 36 , 37 , 38 , 39 , 40 ]. The existing studies summarized in Table 4 are all sepsis predictions, and the AUROC is included in the evaluation criteria; however, the definition of sepsis and the dataset used in the experiment are different.…”

“…We screened several studies that used this dataset and summarized those that produced high AUROC scores. Li et al [ 34 ] recorded an AUROC of 0.75 for a 12 h prediction, and Yang et al [ 36 ] obtained an AUROC of 0.85 for a 1 h prediction. Li et al [ 38 ] proposed a model based on LightGBM and obtained an AUROC of 0.85 for a 6 h prediction.…”

“…The created RNN model was compared with existing scoring systems (SOFA, qSOFA, and SAPS II), machine learning models such as InSight and LSTM, and the results of previous studies [ 14 , 20 , 24 , 30 , 31 , 32 , 33 , 34 , 35 , 36 , 37 , 38 , 39 , 40 ].…”

Artificial Intelligence for Clinical Decision Support in Sepsis

“…In addition to the above-mentioned deep learning, some people have developed an explainable AI model for early prediction of sepsis. They developed a model based on shared ICU public data and verified the challenge score in a completely hidden population ( 19 ). The explainable AI model extracts 168 features per hour and is trained to achieve real-time prediction of sepsis.…”

Clinical factors associated with rapid treatment of sepsis

“…Numerous efforts have been made to improve prognostic accuracy and efficiency for sepsis and its complications via machine learning techniques. For example, Yang et al [8], Komorowski et al [9] and Reyna et al [10] developed artificial intelligence models to predict sepsis in intensive care; while Mao et al [11] and Lauritsen et al [12] extended the prediction application to ED and general ward. Itzhak et al [13] and Cherifa et al [14] developed models to predict acute hypertensive or hypotensive episodes among ICU admissions.…”