AI to read onsite cytology from endoscopic ultrasound fine needle biopsies
Artificial Intelligence for Onsite Cytology Evaluation in Endoscopic Ultrasound
This project will test whether ChatGPT can accurately read cell samples taken during endoscopic ultrasound fine needle biopsies from adults with solid gastrointestinal masses, compared to expert pathologists.
Quick facts
| Study type | Observational |
|---|---|
| Enrollment | 200 (estimated) |
| Ages | 18 Years and up |
| Sex | All |
| Sponsor | Orlando Health, Inc. Academic / other |
| Locations | 1 site (Orlando, Florida) |
| Trial ID | NCT07228689 on ClinicalTrials.gov |
What this trial studies
This observational project will compare interpretations generated by the ChatGPT AI platform with cytopathologist readings of cytology specimens obtained during endoscopic ultrasound-guided fine needle biopsies of solid gastrointestinal masses. Standard-of-care EUS-FNB procedures will collect samples, which will be read by onsite cytopathologists and processed for AI interpretation while blinded to the human result. Measures will include diagnostic concordance, rates of inadequate samples, and classification of abnormalities to determine AI accuracy against the gold standard. No experimental interventions are given beyond routine biopsy and pathological review, and enrollment is limited to adult patients at the Orlando Health Digestive Health Institute.
Who should consider this trial
Good fit: Adults aged 18 or older who are undergoing endoscopic ultrasound-guided fine needle biopsy of a solid mass lesion in the gastrointestinal tract.
Not a fit: Patients under 18, those with cystic gastrointestinal lesions, or people not undergoing EUS-guided sampling of solid masses are not eligible and would not benefit from the findings.
Why it matters
Potential benefit: If successful, this approach could expand access to rapid onsite cytology interpretation, reduce inadequate or repeat biopsies, and speed diagnosis for patients with GI mass lesions.
How similar studies have performed: Other AI and image-analysis tools have shown promise in cytology and rapid onsite evaluation, but applying ChatGPT specifically to cytology image interpretation is novel and not yet widely validated.
Eligibility criteria
Show full inclusion / exclusion criteria
Inclusion Criteria: * Age 18 years old or greater * Patients undergoing EUS-guided fine needle biopsy of solid mass lesions in the gastrointestinal tract Exclusion Criteria: * Age less than 18 years * Patients not undergoing sampling of solid mass lesions at EUS * Patients with cystic lesions in the gastrointestinal tract
Where this trial is running
Orlando, Florida
- Orlando Health Digestive Health Institute — Orlando, Florida, United States (Recruiting)
Study contacts
- Principal investigator: Ji Young Bang, MD MPH — Orlando Health
- Study coordinator: Ji Young Bang, MD MPH
- Email: jiyoung.bang@orlandohealth.com
- Phone: 321.842.2273
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.