Toward Robust AI-Assisted Dietary Assessment for Diabetes Self-Management: Quantifying and Decomposing Large Language Model Prediction Variability From Meal Images
JMIR Diabetes · 2026-09-04 · 11 · e102715-e102715
DOI: 10.2196/102715 ↗At a glance
Journal article in JMIR Diabetes, published online on 2026-09-04. Written by Zhaohua Wang and 2 co-authors. Institution: Department of Health and Behavioral Informatics & Therapeutics (HABIT), Graduate School of Medicine, The University of Tokyo, Faculty of Medicine Bld…. Cited 1 times so far (Crossref); 4 references. The full text is freely available in PubMed Central.
Topics: Diabetes mellitus, Künstliche Intelligenz in der Pflege, Selbstmanagement bei chronischen Erkrankungen
Abstract (original language)
Nutrient estimation from meal images by multimodal large language models can support diabetes self-management, but its robustness is limited by variability driven by both sensitivity to visual presentation and inherent model instability. Accounting for and mitigating this variability is essential for robust AI-assisted dietary assessment.
Subjects
Author keywords
Authors and institutions
- Department of Health and Behavioral Informatics & Therapeutics (HABIT), Graduate School of Medicine, The University of Tokyo, Faculty of Medicine Bldg. 2, 2nd Floor, 7-3-1 Hongo, Bunkyo-ku, Tokyo, 113-0033, Japan, 81 3-…
- Department of Diabetes and Metabolic Diseases, The University of Tokyo Hospital, Tokyo, Japan.
Cite
APA
Wang, Z., Lane, D., & Waki, K. (2026). Toward Robust AI-Assisted Dietary Assessment for Diabetes Self-Management: Quantifying and Decomposing Large Language Model Prediction Variability From Meal Images. JMIR Diabetes, 11, e102715-e102715. https://doi.org/10.2196/102715
Vancouver
Wang Z, Lane D, Waki K. Toward Robust AI-Assisted Dietary Assessment for Diabetes Self-Management: Quantifying and Decomposing Large Language Model Prediction Variability From Meal Images. JMIR Diabetes. 2026;11:e102715-e102715. doi:10.2196/102715
BibTeX
@article{Wang2026,
author = {Wang, Zhaohua and Lane, Daniel and Waki, Kayo},
title = {Toward Robust AI-Assisted Dietary Assessment for Diabetes Self-Management: Quantifying and Decomposing Large Language Model Prediction Variability From Meal Images},
journal = {JMIR Diabetes},
year = {2026},
volume = {11},
pages = {e102715--e102715},
doi = {10.2196/102715}
}
EUVIMED