Performance of an Artificial Intelligence-Powered Smartphone Application in Identifying Gluten-free Foods for Individuals With Celiac Disease

2026-04-01

PRE

Samaan, Jamil S. *

Srinivasan, Nitin

Feldman, Erin

Scarcello, Christine

Fadi S. Samaan

Mansour, Achraf

Fehlman, Jared

Lauren Schlussel

Yeo, Yee Hui

Reddy, Kavya

Tatonetti, Nicholas P.

Pimentel, Mark

Kelly Issokson

Rezaie, Ali

DOI：10.14309/ajg.0000000000003863

Original

Original request for help

Save

Abstract

En 中文

INTRODUCTION:Gluten-free diet (GFD) adherence is critical for individuals with celiac disease but is often hindered by food labeling inconsistencies and limited dietary support.METHODS:We evaluated an AI-powered smartphone application's performance in classifying the GFD compatibility of packaged foods. Compatibility was assessed by gastroenterology-focused registered dietitians.RESULTS:Of 100 products (1,884 ingredients), dietitians classified 64 as GFD-compatible, of which only 37.5% (24/64) displayed a gluten-free label. The application demonstrated 99% and 99.8% accuracy at the product and ingredient level, respectively.DISCUSSION:With rigorous evaluation, AI-powered tools may support dietary adherence for patients with celiac disease.

Keywords:

Celiac disease

gluten-free diet

artificial intelligence

nutrition

large language models

Performance of an Artificial Intelligence-Powered Smartphone Application in Identifying Gluten-free Foods for Individuals With Celiac Disease

Abstract

Journal

Organization

Cited Papers

Citing Papers