Large Language Model Evaluated Stand-alone Attention-Assisted Graph Neural Network with Spatial and Structural Information Interaction for Precise Endoscopic Image Segmentation

Juntong Fan, Shuyi Fan, Debesh Jha IEEE Senior Member , Changsheng Fang, Tieyong Zeng, Hengyong Yu IEEE Fellow , Dayang Wang IEEE Senior Member Department of Mathematics, The Chinese University of Hong Kong, Hong Kong. Department of Electronic and Computer Engineering, University of Massachusetts, L · 2025

Read Large Language Model Evaluated Stand-alone Attention-Assisted Graph Neural Network with Spatial and Structural Information Interaction for Precise Endoscopic Image Segmentation by Juntong Fan, Shuyi Fan, Debesh Jha IEEE Senior Member , Changsheng Fang, Tieyong Zeng, Hengyong Yu IEEE Fellow , Dayang Wang IEEE Senior Member Department of Mathematics, The Chinese University of Hong Kong, Hong Kong. Department of Electronic and Computer Engineering, University of Massachusetts, L (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2508-07028.txt.

Description

Finding polyps in colonoscopy images is a first step toward catching bowel cancer early, but polyps hide well. Juntong Fan and co-authors present a network for outlining them in this 2025 paper. Polyps blend with the gut wall, glare under the scope's light, and fade at their edges. Their system, FOCUS-Med, builds its picture from a graph rather than raw pixels alone.

Two graph modules read the image's space layout and its structural links at once. Attention layers weigh which graph features matter most. Topology cues and connectivity help the model hold a polyp's true boundary. A large language model was used to judge the outlines. Sharper boundaries could mean fewer missed polyps during screening.

Catalog record

CREATORJuntong Fan, Shuyi Fan, Debesh Jha IEEE Senior Member , Changsheng Fang, Tieyong Zeng, Hengyong Yu IEEE Fellow , Dayang Wang IEEE Senior Member Department of Mathematics, The Chinese University of Hong Kong, Hong Kong. Department of Electronic and Computer Engineering, University of Massachusetts, L
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE31 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2508-07028

Browse the full library · About the Tower of Babel