Evaluating Multiple Input Strategies of Large Language Models for Gallbladder Polyps on Ultrasound: Comparative Study

Lin Jiang, Jiaqian Yao, Zebang Yang, Fuqiu Tang, Xin Zheng, Xiaoer Zhang, Xiaoyan Xie, Ming Xu, Tongyi Huang · 2025

Read Evaluating Multiple Input Strategies of Large Language Models for Gallbladder Polyps on Ultrasound: Comparative Study by Lin Jiang, Jiaqian Yao, Zebang Yang, Fuqiu Tang, Xin Zheng, Xiaoer Zhang, Xiaoyan Xie, Ming Xu, Tongyi Huang (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/pmc-12777648.txt.

Description

Lin Jiang and co-authors test language models on gallbladder polyp scans in this 2025 study. Most gallbladder polyps are harmless, but some are adenomas that need surgery. Size-based rules can push patients toward needless operations.

The team gathered ultrasound images and reports for polyps of one centimeter or more with known pathology. Two large multimodal models were tried under three input plans: images alone, feature text, and score-based text. Their calls were compared with radiologists and with the guideline rule. Agreement within and between readers was checked as well. The study weighs whether such models could ease radiologist load and patient worry.

Catalog record

CREATORLin Jiang, Jiaqian Yao, Zebang Yang, Fuqiu Tang, Xin Zheng, Xiaoer Zhang, Xiaoyan Xie, Ming Xu, Tongyi Huang
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE42 KB
VERSION1.0
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC12777648

Browse the full library · About the Tower of Babel