Machine Learning for Intraoperative Bleeding Prediction in Patients Undergoing Surgery: Scoping Review
Shiqiong Yan1, Ping Zhang1, Wanwan Qiao2
1Department of Nursing, Sichuan Provincial People's Hospital, University of Electronic Science and Technology of China, No. 32, West Section 2, First Ring Road, Qingyang District, Chengdu, 610072, China, 86 028-87393999, 86 028-87393999.
JMIR Medical Informatics
|June 10, 2026
Summary
Machine learning models show promise for predicting intraoperative bleeding, but methodological flaws and lack of external validation limit clinical use. Future research needs prospective validation and transparent reporting for reliable implementation.
Area of Science:
- Medical Informatics
- Artificial Intelligence in Medicine
- Surgical Safety
Background:
- Intraoperative bleeding is a critical surgical event impacting patient outcomes.
- Machine learning (ML) models offer potential for predicting intraoperative bleeding.
- Methodological rigor and clinical translation of ML models face significant challenges.
Conclusions:
- ML models show technical promise but suffer from high risk of bias and limited external validation, restricting clinical reliability.
- Lack of transparency and poor reporting further impede clinical utility.
- Future research requires prospective multicenter validation, TRIPOD adherence, and improved interpretability for clinical integration.
