Using AI to interpret ROTEM blood-clot tests during cardiac and liver transplant surgery
Assessment of Artificial Intelligence Algorithms for ROTEM Analysis in Coagulation Management
This project will test whether AI models can read intraoperative ROTEM results and suggest appropriate treatments for adults having elective cardiac or liver transplant surgery.
Quick facts
| Study type | Observational |
|---|---|
| Enrollment | 144 (estimated) |
| Ages | 18 Years and up |
| Sex | All |
| Sponsor | Ondokuz Mayıs University Academic / other |
| Locations | 2 sites (Istanbul and 1 other locations) |
| Trial ID | NCT07043556 on ClinicalTrials.gov |
What this trial studies
This prospective observational validation will convert de-identified intraoperative ROTEM results into structured clinical vignettes and prompt multiple AI language models with standardized ROTEM parameters. At least three experienced clinicians will independently interpret each vignette to form a gold-standard consensus for diagnosis of coagulopathy type and recommended treatments. AI outputs (e.g., ChatGPT, Gemini) will be compared to the expert panel for diagnostic concordance and agreement on treatment recommendations, using established ROTEM interpretation algorithms as background context. The study includes adult elective cardiac surgery and liver transplant cases with complete, technically valid ROTEM panels.
Who should consider this trial
Good fit: Adults (≥18 years) undergoing elective cardiac surgery (CABG, valve, aortic procedures) or liver transplantation with complete intraoperative ROTEM panels and who provide informed consent are ideal candidates.
Not a fit: Pediatric patients, emergency or redo surgeries, those without complete or technically valid ROTEM data, and patients who do not undergo the listed elective procedures are unlikely to benefit from participation.
Why it matters
Potential benefit: If successful, AI could speed and standardize ROTEM interpretation in the operating room, helping clinicians make timely transfusion and hemostatic treatment decisions that may reduce bleeding and unnecessary blood product use.
How similar studies have performed: ROTEM-guided transfusion algorithms have improved outcomes and some machine-learning approaches have shown promise for viscoelastic data, but using large language models like ChatGPT or Gemini for ROTEM interpretation is a relatively novel approach that is not yet proven.
Eligibility criteria
Show full inclusion / exclusion criteria
Inclusion Criteria: * Adult patients undergoing elective cardiac surgery (CABG, valve surgery, aortic procedures) * Adult patients undergoing liver transplantation * Availability of complete ROTEM results (EXTEM, INTEM, FIBTEM, +/- HEPTEM, APTEM) * Informed written consent obtained Exclusion Criteria: * Incomplete or technically invalid ROTEM data * Pediatric patients (\<18 years) * Refusal to participate or lack of informed consent * Emergency and redo surgeries
Where this trial is running
Istanbul and 1 other locations
- İstanbul Aydın Üniversitesi Sağlık Uygulama ve Araştırma Merkezi Medical Park Florya Hastanesi — Istanbul, Turkey (Türkiye) (Recruiting)
- Ondokuz Mayis University — Samsun, Turkey (Türkiye) (Recruiting)
Study contacts
- Principal investigator: Burhan Dost, Assoc. Prof. — Ondokuz Mayıs University
- Study coordinator: Burhan Dost, Assoc. Prof.
- Email: burhandost@hotmail.com
- Phone: +90 532 704 24 93
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.