Related Experiment Video
Updated: Aug 17, 2025
Predicting Amputation using Local Circulating Mononuclear Progenitor Cells in Angioplasty-treated Patients with Critical Limb Ischemia
Published on: September 22, 2020
Early predicting 30-day mortality in sepsis in MIMIC-III by an artificial neural networks model
Yingjie Su1, Cuirong Guo1, Shifang Zhou1
1Department of Emergency Medicine, The Affiliated Changsha Central Hospital, Hengyang Medical School, University of South China, NO. 161 Shaoshan South Road, Changsha, 410004, Hunan, China.
Objective:
Early identifying sepsis patients who had higher risk of poor prognosis was extremely important. The aim of this study was to develop an artificial neural networks (ANN) model for early predicting clinical outcomes in sepsis.
Methods:
This study was a retrospective design. Sepsis patients from the Medical Information Mart for Intensive Care-III (MIMIC-III) database were enrolled. A predictive model for predicting 30-day morality in sepsis was performed based on the ANN approach.
Results:
A total of 2874 patients with sepsis were included and 30-day mortality was 29.8%. The study population was categorized into the training set (n = 1698) and validation set (n = 1176) based on the ratio of 6:4. 11 variables which showed significant differences between survivor group and nonsurvivor group in training set were selected for constructing the ANN model. In training set, the predictive performance based on the area under the receiver-operating characteristic curve (AUC) were 0.873 for ANN model, 0.720 for logistic regression, 0.629 for APACHEII score and 0.619 for SOFA score. In validation set, the AUCs of ANN, logistic regression, APAHCEII score, and SOFA score were 0.811, 0.752, 0.607, and 0.628, respectively.
Conclusion:
An ANN model for predicting 30-day mortality in sepsis was performed. Our predictive model can be beneficial for early detection of patients with higher risk of poor prognosis.
Insights
An artificial neural network (ANN) model effectively predicts 30-day mortality in sepsis patients. This tool aids in early identification of individuals at high risk for poor prognosis, improving sepsis management.
Area of Science:
- Medical Informatics
- Artificial Intelligence in Medicine
- Critical Care Medicine
Background:
- Early identification of sepsis patients at high risk of mortality is crucial for timely intervention.
- Existing predictive models may lack the accuracy needed for effective early risk stratification in sepsis.
Purpose of the Study:
- To develop and validate an artificial neural network (ANN) model for early prediction of clinical outcomes in sepsis patients.
- To assess the predictive performance of the ANN model compared to traditional scoring systems.
Main Methods:
- Retrospective analysis of sepsis patients from the Medical Information Mart for Intensive Care-III (MIMIC-III) database.
- Development of an ANN model using 11 significant variables identified from training data to predict 30-day mortality.
- Validation of the ANN model against logistic regression, APACHE II, and SOFA scores.
Main Results:
- The study included 2874 sepsis patients with a 30-day mortality rate of 29.8%.
- The ANN model achieved an Area Under the Receiver Operating Characteristic Curve (AUC) of 0.873 in the training set and 0.811 in the validation set.
- The ANN model demonstrated superior predictive performance compared to logistic regression, APACHE II, and SOFA scores in both training and validation sets.
Conclusions:
- An ANN model was successfully developed for predicting 30-day mortality in sepsis.
- This predictive model offers a valuable tool for the early detection of sepsis patients with a higher risk of poor prognosis.
