Artificial Intelligence-Guided Detection of Blood Vessels to Enhance Safety in Third-Space Endoscopic Procedures

Asian Institute of Gastroenterology, India

Status: Enrolling

Conditions: Achalasia Cardia; Tumor

Treatments: Device: AI generated segmentation mask for sub-mucosal blood vessels

Study type: Interventional

Funder types: Other

Identifiers

NCT07399652

AITSE-1

Details and patient eligibility

About

This prospective study aims to evaluate the performance of a novel Artificial Intelligence (AI) clinical decision support tool during third space endoscopic procedures, such as Endoscopic Submucosal Dissection (ESD) and Peroral Endoscopic Myotomy (POEM).

While these procedures are effective for treating gastrointestinal neoplasms and motility disorders, they carry risks of intraprocedural bleeding and perforation if submucosal blood vessels are not correctly identified and coagulated. Building on previous retrospective validation, this study will assess whether a real-time artificial intelligence model can assist endoscopists in detecting and delineating blood vessels more accurately and faster during live human procedures.

Full description

Background and Rationale

Third-space endoscopy procedures are technically demanding. The primary challenge lies in the inadvertent transection of submucosal vessels, which leads to bleeding that obscures the surgical field and increases the risk of perforation. Currently, vessel identification is entirely operator-dependent.

Our team has developed a deep-learning based artificial intelligence model trained on 250,000 annotated images from 150 POEM procedures. This model is optimized for minimal latency, allowing for real-time visual overlays (delineation) of blood vessels on the endoscopic monitor.

Study Objectives The primary objective is to evaluate the Vessel Detection Rate (VDR)-the proportion of vessels identified by the endoscopist when assisted by the AI compared to standard practice.

The study will also investigate:

Vessel Detection Time (VDT): The latency between a vessel appearing in the field of view and its identification.

Study Design & Workflow:

In this prospective study, the AI system will be integrated into the Olympus EVIS X1 series endoscopy stack. As the endoscopist navigates the submucosal space, the AI will provide real-time visual segmentation masks highlighting vessels. The performance will be recorded and compared against a post-procedure review by independent experts to calculate sensitivity and detection speed.

Enrollment: 20 estimated patients

Sex: All

Volunteers: No Healthy Volunteers

Inclusion criteria

• Patients diagnosed with Achalasia Cardia or neoplasms.

Exclusion criteria

• Patients with conditions deemed unsuitable for third space endoscopy procedures (e.g.: Candidiasis)

Trial design

Primary purpose: Device Feasibility

Allocation: Randomized

Interventional model: Parallel Assignment

Masking: None (Open label)

---

20 participants in 2 patient groups

With Artificial Intelligence

Active Comparator group

Description:

Endoscopist will see the AI generated segmentation mask

Treatment:

Device: AI generated segmentation mask for sub-mucosal blood vessels

Without Artificial Intelligence

No Intervention group

Description:

Endoscopist will not see the AI generated segmentation mask