Key Summary:
- AI improved nutritional management for IBD in 16 studies.
- AI predicted treatment response and answered nutritional questions with 83.0% accuracy.
- Larger studies are needed before AI can support routine IBD care.
AI demonstrated promising applications in the nutritional management of inflammatory bowel disease (IBD), with evidence suggesting benefits in dietary pattern recognition, treatment response prediction, personalised patient support, and identification of patient needs, according to a recent scoping review.
AI Supported Nutritional Management
Diet plays an important role in the onset, progression, and prognosis of IBD. However, nutritional management remains challenging because clear dietary guidance is limited. To evaluate the emerging role of AI in this area, researchers conducted a scoping review following the Preferred Reporting Items for Systematic Reviews and Meta Analyses Extension for Scoping Reviews checklist. A systematic search of 11 databases examined studies published from database inception until February 2026 that investigated AI applications for nutritional management in patients with IBD.
From 4,560 records identified, 16 studies met the inclusion criteria. Across these studies, AI technologies included traditional machine learning, deep learning, natural language processing, and multi omics integrated analysis.
AI Delivered Multiple Clinical Applications
The review found that AI supported several aspects of nutritional management. Machine learning approaches identified dietary patterns associated with lower inflammation risk, including plant-based diets. Predictive models also estimated responses to treatment, including the success of total parenteral nutrition. One machine learning model achieved 90% accuracy when distinguishing Crohn’s disease from ulcerative colitis.
AI also enhanced personalised information support. Conversational AI tools answered nutritional questions with 83.0% accuracy, while smartphone applications influenced dietary behaviours among patients with IBD. In addition, natural language processing and Latent Dirichlet Allocation topic modelling identified common patient concerns, including treatment experiences, dietary advice, and psychological burden.
The review also reported preliminary evidence that AI applications reduced inflammatory markers and improved the gut microbiota. Furthermore, faecal metabolites were identified as reliable indicators of disease.
Further Validation Remained Essential
Although the findings highlighted the potential of AI to improve nutritional management in IBD, the researchers noted important limitations. Many studies included small sample sizes, limiting the generalisability of the findings, while the available evidence remained preliminary and heterogeneous.
The authors concluded that larger scale studies and multidisciplinary collaboration will be essential to move AI beyond proof of concept and towards routine clinical practice in the nutritional management of IBD.
Reference
Qian X et al. Artificial Intelligence in the nutritional management of inflammatory bowel disease: a scoping review. J Multidiscip Healthc. 2026;19:614374.
Featured Image: Prathankarnpap on Adobe Stock
Each article is made available under the terms of the Creative Commons Attribution-Non Commercial 4.0 License.
