Related Experiment Video
Updated: Jun 22, 2026

Flexible Colonoscopy in Mice to Evaluate the Severity of Colitis and Colorectal Tumors Using a Validated Endoscopic Scoring System
Published on: October 16, 2013
Reading systemic disease from the feline fundus: A multicenter comparison of multimodal large language models and
Emre Eren1, Murat İlgün2, Büşra Kibar Kurt3
1Department of Internal Medicine, Faculty of Veterinary Medicine, Atatürk University, Erzurum, Turkey.
Abstract:
This multicenter observer-performance study evaluated whether systemic diseases in cats can be inferred from fundus images and structured fundus descriptors and compared the diagnostic performance of large language models (LLMs) with experienced and novice veterinary clinicians. A total of 50 feline cases with representative fundus photographs and harmonized descriptors were retrospectively collected from three centers in Türkiye. Each case was assigned a single canonical diagnosis by a masked expert panel. Two LLMs (ChatGPT-5, ChatGPT-5 Thinking) are tested in text-only and text and image (multimodal) modes and compared with two expert ophthalmologists and two novice clinicians. The primary endpoint was an ordinal diagnostic score (0-3), while secondary endpoints included Top-1/Top-2/Top-3 accuracy, inter-reader agreement, and response times. Multimodal LLMs achieved near-expert diagnostic performance, with Top-1, Top-2, and Top-3 accuracies ranging from 34 % to 46 %, 62-74 %, and 74-86 %, respectively. Multimodal LLMs performed significantly better than novice readers (p < 0.05), whereas no significant differences were found between LLMs and expert clinicians. Response times were substantially shorter for LLMs (13.7-84.6 s) compared to humans (117-132 s), indicating potential workflow efficiency gains. These findings suggest that multimodal LLMs can provide rapid, near-expert-level diagnostic support in interpreting feline fundus images for systemic disease screening. Incorporating LLM assisted analysis into veterinary ophthalmology workflows could enhance diagnostic efficiency without compromising clinician oversight.
More Related Videos
Related Concept Videos
Mouse Models of Cancer Study
The development of transgenic, knockout, and knock-in mice has led to an exponential increase in their use as model organisms in research,...
Classification of Illness
An illness is a response to a disease in which the person's level of functioning is changed compared with a previous level. The general classification of illness includes acute and chronic.
Acute illness is severe and...
Mouse Models of Cancer Study
The development of transgenic, knockout, and knock-in mice has led to an exponential increase in their use as model organisms in research,...
Methods of Documentation VI: Case Management Model
For example, a patient with a chronic illness...
Introduction to Language of Pathophysiology l
Introduction to Language of Pathophysiology ll

