Using AI to help doctors find and outline blood vessels during POEM and ESD
Artificial Intelligence-Guided Detection of Anatomical Markers to Enhance Safety in Third-Space Endoscopic Procedures
This study will test whether an AI tool can help endoscopists spot and outline submucosal blood vessels faster and more accurately during third-space procedures like POEM and ESD for people with achalasia or gastrointestinal tumors.
Quick facts
| Phase | Not applicable |
|---|---|
| Study type | Interventional |
| Enrollment | 20 (estimated) |
| Sex | All |
| Sponsor | Asian Institute of Gastroenterology, India Academic / other |
| Locations | 1 site (Hyderabad, Telangana) |
| Trial ID | NCT07399652 on ClinicalTrials.gov |
What this trial studies
This is a prospective, single-center study that integrates a deep-learning AI model into the Olympus EVIS X1 endoscopy stack to generate real-time segmentation overlays of submucosal blood vessels during third-space procedures. The model was trained on 250,000 annotated images from prior POEM procedures and is optimized for minimal latency so overlays appear live on the endoscopic monitor. Investigators will record the AI-assisted vessel detection performance during procedures and compare Vessel Detection Rate and Vessel Detection Time to a post-procedure expert review to calculate sensitivity and detection speed. The goal is to determine whether real-time AI guidance improves how quickly and accurately vessels are identified, potentially reducing intraprocedural bleeding and perforation risk.
Who should consider this trial
Good fit: Ideal candidates are patients scheduled for third-space endoscopic procedures (POEM, ESD, STER) for achalasia cardia or gastrointestinal neoplasms at the enrolling center.
Not a fit: Patients who are not candidates for third-space endoscopy or who have conditions that make these procedures unsuitable (for example active mucosal infection such as candidiasis) are unlikely to benefit from this intervention.
Why it matters
Potential benefit: If successful, the AI tool could reduce bleeding and perforation by helping clinicians identify and coagulate submucosal vessels earlier and more reliably during third-space endoscopy.
How similar studies have performed: Retrospective validation and preclinical work of this model have shown promise, but prospective real-time use in live human procedures remains novel.
Eligibility criteria
Show full inclusion / exclusion criteria
Inclusion Criteria: * Patients diagnosed with Achalasia Cardia or neoplasms. Exclusion Criteria: * Patients with conditions deemed unsuitable for third space endoscopy procedures (e.g.: Candidiasis)
Where this trial is running
Hyderabad, Telangana
- Asian Institute of Gastroenterology — Hyderabad, Telangana, India (Recruiting)
Study contacts
- Principal investigator: Abhishek Tyagi — Asian Intitute of Gastroenterology
- Study coordinator: Abhishek Tyagi, M.S.
- Email: mr.tyagi@gmail.com
- Phone: 919989154556
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.