Prediction of Deterioration in Critically Ill Patients with Heart Failure Based on Vital Signs Monitoring
- Resource Type
- Conference
- Authors
- Zhang, Shengyu; Yang, Kang; Ye, Wenyu; Jiang, Haoyu; He, Xianliang; Wang, Lei; Li, Yijing
- Source
- 2022 Computing in Cardiology (CinC) Computing in Cardiology (CinC), 2022. 498:1-4 Sep, 2022
- Subject
- Bioengineering
Computing and Processing
Signal Processing and Analysis
Databases
Current measurement
Medical services
Predictive models
Prediction algorithms
Feature extraction
Real-time systems
- Language
- ISSN
- 2325-887X
This study aims to develop a real-time machine learning model for acute heart failure onset based on vital signs in bedside monitoring. A group of 2284 patients were analyzed retrospectively from the Medical Information Mart for Intensive Care III database. We extracted various features building machine learning model. Extreme Gradient Boosting was used to develop the real-time prediction model. The validation on test set gave decent early warning performance. The model prediction can provide more timely notifications for doctors to perform better treatment for patients.