Predicting 30-day heart problems after gastrointestinal surgery using AI versus standard risk scores
Value of Some Risk Scores in Predicting Cardiovascular Events After Gastrointestinal Surgery
This project will test whether an AI-based multimodal model can better predict major heart problems within 30 days after gastrointestinal surgery for adults at Bach Mai Hospital than commonly used risk scores.
Quick facts
| Study type | Observational |
|---|---|
| Enrollment | 5000 (estimated) |
| Ages | 16 Years and up |
| Sex | All |
| Sponsor | Bach Mai Hospital Academic / other |
| Locations | 1 site (Hà Nội) |
| Trial ID | NCT07539532 on ClinicalTrials.gov |
What this trial studies
This is a mixed retrospective–prospective observational effort at Bach Mai Hospital to develop an AI-based multimodal prediction model for major cardiovascular events occurring within 30 days after gastrointestinal surgery. The model will draw on preoperative, intraoperative, and postoperative clinical, laboratory, physiologic, and procedural data from adult patients treated in 2025–2026. Investigators will compare the AI model's predictive performance with traditional scores (RCRI, ACS NSQIP MICA, and ACS NSQIP SRC) using measures such as discrimination, calibration, and clinical usefulness. The primary outcome is major cardiovascular events (cardiovascular death, nonfatal MI, cardiac arrest with return of spontaneous circulation, new stroke, and related events) within 30 days after surgery.
Who should consider this trial
Good fit: Adult patients (age ≥18) undergoing gastrointestinal surgery at Bach Mai Hospital between January 2025 and December 2026 with sufficient preoperative, intraoperative, and postoperative data are the intended participants.
Not a fit: Patients without the required clinical data, those who die within 24 hours from a clearly non-cardiovascular cause, pediatric patients, or those undergoing non-gastrointestinal procedures are unlikely to benefit from this model.
Why it matters
Potential benefit: If successful, the AI model could more accurately identify patients at high risk of early postoperative cardiovascular events, enabling closer monitoring and targeted preventive care.
How similar studies have performed: Some prior machine-learning models have improved prediction of surgical and cardiac outcomes, but multimodal AI models with robust external or prospective validation remain limited.
Eligibility criteria
Show full inclusion / exclusion criteria
Inclusion Criteria: * Adults aged 18 years or older. * Undergoing gastrointestinal surgery at Bach Mai Hospital between January 2025 and December 2026. * Available preoperative, intraoperative, and postoperative data sufficient for analysis. Exclusion Criteria: * Death within 24 hours after surgery due to a clearly non-cardiovascular cause. * Incomplete data required for analysis.
Where this trial is running
Hà Nội
- Bach Mai hospital — Hà Nội, Vietnam (Recruiting)
How to participate
- Review the eligibility criteria above with your treating physician.
- Visit the official trial page on ClinicalTrials.gov for the most current contact information and recruitment status.
- Contact the listed study coordinator or principal investigator to request pre-screening. Pre-screening is free and never obligates you to enroll.