Machine Learning-Based Early Prediction of Sepsis Using Electronic Health Records: A Systematic Review
Problem
Detecting sepsis early enough to intervene, since it rapidly progresses to life-threatening forms; ML/DL on electronic health records was surveyed as an approach.
Why it matters
Sepsis is life-threatening and timely detection directly improves patient survival.
What the AI did
This is a systematic review of 42 studies that used machine learning and deep learning on EHRs to detect or early-predict sepsis in adults; no new AI system was proposed.
Results
1942 articles screened; 42 studies selected, predominantly retrospective (n=38), across diverse geographic settings.
Impact
Synthesizes the evidence base for ML-based sepsis prediction; no clinical deployment or performance benchmark is stated.
Record
| Venue | Journal of Clinical Medicine |
|---|---|
| Published | 2023 |
| Work type | review |
| Authors | Khandaker Reajul Islam, Johayra Prithula, Jaya Kumar, Toh Leong Tan, Mamun Bin Ibne Reaz, Md. Shaheenur Islam Sumon, Muhammad E. H. Chowdhury |
| Citations | 118 (OpenAlex, at import time) |
| References | 81 |
| Model or method | not stated in source |
| Organization | not stated in source |
| Primary source | https://doi.org/10.3390/jcm12175658 |
| Open access | gold |
| Keywords | sepsis, machine learning, electronic health records, early prediction, systematic review |
Importance rating
major — millions of lives, an industry or a field
Abstract states sepsis is life-threatening with significant global health impacts and that timely detection improves outcomes; early prediction addresses a major clinical problem.