Related Experiment Video
Updated: Jun 1, 2025

Laparoscopy-endoscopy Cooperative Surgery for the Treatment of Gastric Gastrointestinal Stromal Tumors
Published on: February 19, 2022
GastroHUN an Endoscopy Dataset of Complete Systematic Screening Protocol for the Stomach
Diego Bravo1,2, Juan Frias3,4, Felipe Vera3,4
1Universidad Nacional de Colombia, Bogotá, 1100111, Colombia. dbravoh@unal.edu.co.
GastroHUN is a new open dataset for stomach screening, featuring 8,834 images and 4,729 labeled sequences. This resource aids AI development for improved gastrointestinal disease detection using endoscopy.
Area of Science:
- Medical imaging and diagnostics
- Artificial intelligence in healthcare
- Gastroenterology research
Background:
- Endoscopy is crucial for gastrointestinal disease diagnosis, with systematic protocols enhancing early detection of premalignant conditions.
- Existing endoscopy image datasets face challenges in labeling consistency and accessibility, hindering AI model development and generalizability.
- Accurate identification of upper gastrointestinal anatomical landmarks is essential for precise endoscopic procedures.
Purpose of the Study:
- To introduce GastroHUN, a novel, open-access dataset for stomach screening endoscopy.
- To provide a comprehensive resource for developing and benchmarking AI models for gastrointestinal endoscopy.
- To address limitations in existing datasets regarding labeling consistency and anatomical landmark annotation.
Main Methods:
- Development of GastroHUN dataset comprising 8,834 images and 4,729 labeled video sequences from 387 patients.
- Annotation of the dataset by four experts, covering 22 stomach anatomical landmarks and including unqualified images.
- Establishment of baseline deep learning models for image and sequence classification using the GastroHUN dataset.
Main Results:
- GastroHUN dataset offers a large-scale, systematically collected resource for AI in endoscopy.
- The dataset includes detailed annotations of 22 anatomical landmarks, improving precision in AI model training.
- Baseline deep learning models demonstrate the utility of GastroHUN for AI development in gastrointestinal imaging.
Conclusions:
- GastroHUN provides a valuable, open-access benchmark for AI research in upper gastrointestinal endoscopy.
- The dataset facilitates the development of more accurate and generalizable AI algorithms for detecting gastrointestinal diseases.
- Standardized protocols and expert annotations in GastroHUN enhance its utility for advancing AI-driven endoscopic diagnostics.
Related Concept Videos
Endoscopic Procedures I: Esophagogastroduodenoscopy
During an EGD, the endoscope can be used to:
Endoscopic Procedures II: Colonoscopy
Endoscopic Procedures IV: Sigmoidoscopy and Laproscopy
Sigmoidoscopy
Sigmoidoscopy is a diagnostic procedure that uses a flexible sigmoidoscope equipped with a light source and camera to examine the rectum and sigmoid colon. The procedure involves inserting the tube through the anus...
Endoscopic Procedures III: Video Capsule Endoscopy
Assessment of the Gastrointestinal System I: Subjective Data
Health History
The initial step in assessing the GI system is obtaining a comprehensive health history. This includes inquiring about the patient's history or presence of problems...
Peptic Ulcer Disease III: Clinical Manifestations and Diagnostic Studies
Few clinical manifestations differentiate gastric ulcers from duodenal ulcers. Distinctions in the location, timing, and pain relief are crucial for healthcare providers in differentiating between gastric and duodenal ulcers during clinical assessments.

