Using AI to Customize Genomic-Based Nutrition for Rehab Patients
Keywords:
Artificial Intelligence, Dietary Adherence, Genomic Medicine, Metabolic Syndrome, Nutritional Genomics, Personalized Nutrition, Rehabilitation TherapyAbstract
Background: Personalized nutrition is gaining prominence in clinical rehabilitation due to its ability to tailor dietary recommendations based on genetic, metabolic, and behavioral factors. Traditional, generalized nutritional guidelines often fall short in addressing individual variability, especially in recovery-focused populations. Advances in artificial intelligence (AI) and nutrigenomics present an opportunity to optimize dietary interventions through data-driven customization. However, the real-world clinical application of AI-driven genomic nutrition in rehabilitation, particularly within South Asian healthcare contexts, remains insufficiently explored.
Objective: To assess the impact of AI-based, genome-informed dietary planning on metabolic, inflammatory, behavioral, and subjective recovery outcomes among rehabilitation patients in Punjab, Pakistan.
Methods: A prospective, mixed-methods study was conducted between January and May 2025 at two rehabilitation centers in Punjab. A total of 44 adults (aged 25–65) were purposively assigned to either an AI-assisted genomic nutrition group (n=23) or a control group receiving standard dietary guidance (n=21). Genomic data were collected using SNP genotyping (including MTHFR, FTO, CYP1A2 variants). Personalized dietary plans were generated through AI algorithms incorporating machine learning models (Random Forest, SVM), and delivered via a mobile application with bi-weekly monitoring. Anthropometric, biochemical, and subjective data were collected at baseline and post-intervention. Outcomes were analyzed using paired t-tests and ANCOVA with p<0.05 considered significant.
Results: The AI group showed a significant decrease in BMI (27.6 to 25.8 kg/m²), waist circumference (92.3 to 88.7 cm), CRP levels (4.2 to 2.6 mg/L), and fasting glucose (102.5 to 94.1 mg/dL), compared to minimal changes in the control group. Total cholesterol declined by 14.1 mg/dL in the AI group versus 2.4 mg/dL in controls. Dietary adherence was higher in the AI group (89.5% vs. 61.7%), along with increased app usage consistency (93.1% vs. 58.4%), self-reported energy (8.2 vs. 6.3), and recovery satisfaction (8.7 vs. 6.5) on a 10-point scale.
Conclusion: AI-integrated, genomic-based nutrition significantly improved both objective health metrics and subjective rehabilitation outcomes compared to standard dietary practices. These findings suggest that such technology can be feasibly and effectively implemented in rehabilitation settings, even in resource-constrained regions.
References
1. Di Renzo L, Gualtieri P, Romano L, Marrone G, Noce A, Pujia A, et al. Role of personalized nutrition in chronic-degenerative diseases. 2019;11(8):1707.
2. Dainis AM, Ashley EAJJBtTS. Cardiovascular precision medicine in the genomics era. 2018;3(2):313-26.
3. Kohlmeier M, Chirita A, Beckett E, Angelino D, Del Rio D, Niculescu M. 13th Congress of the International Society of Nutrigenetics/Nutrigenomics (ISNN). 2019.
4. DHABI AJOjotWUoWHS. WUWHS. 2018;27(3).
5. Gadde KM, Martin CK, Berthoud H-R, Heymsfield SBJJotACoC. Obesity: pathophysiology and management. 2018;71(1):69-84.
6. Bell SC, Mall MA, Gutierrez H, Macek M, Madge S, Davies JC, et al. The lancet respiratory medicine commission on the future of care of cystic fibrosis. 2019;8(1):65.
7. Jianming J. List of research projects available for prospective graduate students.
8. Verma M, Hontecillas R, Tubau-Juni N, Abedi V, Bassaganya-Riera JJFiN. Challenges in personalized nutrition and health. Frontiers Media SA; 2018. p. 117.
9. Hu G-M, Lee VD, Lin H-Y, Mao P-W, Liu H-Y, Peh J-H, et al. Single-cell technologies for cancer therapy. Handbook of Single Cell Technologies: Springer; 2019. p. 1-84.
10. Dutton JS, Hinman SS, Kim R, Wang Y, Allbritton NLJTib. Primary cell-derived intestinal models: recapitulating physiology. 2019;37(7):744-60.
11. Hedayat KM, Lapraz J-C. The theory of Endobiogeny: Volume 1: Global systems thinking and biological modeling for clinical medicine: Academic Press; 2019.
12. Barker R, Barretto M, Davies P, Goetz CG, Hague T, Hattori N, et al. 5 th World Parkinson Congress 2019 Committee Members.
13. García MS. Facultad de Farmacia y Nutrición.
14. Tanaka D, Inagaki N. Comprehensive whole exome sequencing in Japanese with young-onset diabetes yields insights into their genetic background.
15. Randine P, Muzny M, Micucci D, Arsand EJDT, THERAPEUTICS. System for automatic estimation and delivery of quickly-absorbable carbohydrates. 2019;21(S1):A113-A.
16. Antonucci L, Pergola G, Dwyer D, Torretta S, Romano R, Gelao B, et al. O5. Classification of Schizophrenia Using Machine Learning With Multimodal Markers. 2019;85(10):S107-S.
17. Ma J-Q, Chen L, Wang X, Hao X, Wang L, Yang Y, et al. Global tea science: current status and future needs: Burleigh Dodds Science Publishing; 2018.
18. Khaidakov M. A Pessimistic Guide to Anti-Aging Research: Death is Immortal: Cambridge Scholars Publishing; 2019.
19. Zufall FJCS. XXVIIIth Annual Meeting of the European Chemoreception Research Organization, ECRO 2018. 2019;44:e1-e65.
20. Knight C. Proceedings of the Fifth Dairycare Conference 2018, Thessaloniki, Greece, March 19th and 20th 2018. 2018.
21. Caron P, Broin M, Delaporte E, Duru M, Izopet J, Paul M, et al. Global health. People, animals, plants, the environment: towards an integrated approach to health. Agropolis; 2019.
22. Sutherland M. The Relationship Between fMRI and Symptoms of Major Depressive Disorder (CAN-BIND12): Queen's University (Canada); 2019.
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Farah Niaz Awan, Ali Hamza Arshad (Author)
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
