A multimodal artificial intelligence framework for the early detection of diet-related maculopathy using trophic biomarkers and fundus imaging

Akhilesh Kumar; Deepshikha; Ayub Ali; Yogita Nagar

doi:10.18203/2320-6012.ijrms20260239

Authors

Akhilesh Kumar Department of Information Technology, Santosh Deemed to be University, Ghaziabad, Uttar Pradesh, India https://orcid.org/0009-0008-0882-1539
Deepshikha Department of Clinical, Nutrition and Dietetics, Santosh Deemed to be University, Ghaziabad, Uttar Pradesh, India https://orcid.org/0009-0001-1024-6659
Ayub Ali Department of Optometry, Santosh Deemed to be University, Ghaziabad, Uttar Pradesh, India
Yogita Nagar Department of Clinical, Nutrition and Dietetics, Santosh Deemed to be University, Ghaziabad, Uttar Pradesh, India

DOI:

https://doi.org/10.18203/2320-6012.ijrms20260239

Keywords:

Precision nutrition, Screening, Random Forest, Convolutional neural network, Diet-related maculopathy, Fundus imaging, Trophic biomarkers, Artificial intelligence

Abstract

Background: Diet-related maculopathy (DRM) is a progressive retinal condition associated with chronic nutritional deficiencies, oxidative stress, and impaired macular metabolism. Early detection is challenging because biochemical deterioration precedes clinically visible retinal changes. This study aimed to develop and validate a multimodal artificial intelligence (AI) framework integrating fundus imaging with trophic biomarkers to enhance early DRM detection.

Methods: A prospective diagnostic model validation study was conducted among 580 adults aged 30-65 years at a Asim eye care Center in Ghaziabad, Delhi-NCR between January 2023 and March 2025. Fundus images were analysed using a fine-tuned ResNet-50 convolutional neural network, whereas plasma concentrations of lutein, zeaxanthin, vitamins A, C, and E, zinc, and docosahexaenoic acid (DHA) were processed using a Random Forest classifier. A fusion architecture integrated both outputs. Model performance was assessed through accuracy, sensitivity, specificity, and area under the ROC curve. Longitudinal follow-up assessed predictive lead time.

Results: The multimodal AI model achieved an accuracy of 95.2%, sensitivity of 93.1%, specificity of 96.4%, and an AUC of 0.972. Lutein, zeaxanthin, and DHA were the most significant biochemical predictors, whereas macular reflectance patterns and early drusen signatures were the strongest image-derived features. The model detected DRM on average 11.2 months before clinical diagnosis. Nutritional insufficiencies were present in 42.1% of participants.

Conclusions: The multimodal AI framework demonstrated excellent diagnostic capability and substantial predictive lead time, enabling early identification of DRM and supporting personalized nutritional intervention. This integrative approach may improve preventive retinal care and reduce long-term visual impairment.

Metrics

Metrics Loading ...

References

Beatty S, Koh H, Phil M, Henson D, Boulton M. The role of oxidative stress in the pathogenesis of age-related macular degeneration. Surv Ophthalmol. 2000;45(2):115-34. DOI: https://doi.org/10.1016/S0039-6257(00)00140-5

Chiu C, Taylor A. Nutritional antioxidants and age-related cataract and maculopathy. Exp Eye Res. 2007;84(2):229-45. DOI: https://doi.org/10.1016/j.exer.2006.05.015

Roberts RL, Green J, Lewis B. Lutein and zeaxanthin in eye and skin health. Clin Dermatol. 2009;27(2):195-201. DOI: https://doi.org/10.1016/j.clindermatol.2008.01.011

Ting DSJ, Foo VHX, Yang WYL, Sia JT, Ang M, Lin H, et al. Artificial intelligence in ophthalmology: applications and future directions. Brit J Ophthalmol. 2001;105(2):158-65. DOI: https://doi.org/10.1136/bjophthalmol-2019-315651

Nadeem MW, Goh HG, Hussain M, Liew S, Andonovic I, Khan MA. Deep learning for diabetic retinopathy detection and diagnosis. Sensors (Basel). 2022;22(18):6780. DOI: https://doi.org/10.3390/s22186780

Klein R. Association of carotenoids with macular pigment optical density. Am J Clin Nutr. 2018;108(4):911-21.

Burlina PM. AI-based automated detection of age-related macular degeneration: a review. Progr Retin Eye Res. 2022;86:100973.

De Fauw J. Clinically applicable deep learning for retinal disease diagnosis. Nature Med. 2018;24(9):1342-50. DOI: https://doi.org/10.1038/s41591-018-0107-6

Breiman L. Random forests. Machine Learn. 2001;45(1):5-32. DOI: https://doi.org/10.1023/A:1010933404324

Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition. Arxiv Preprint. 2014.

Li J. Integrative machine learning in ophthalmic imaging. J Translat Med. 2022;20:452.

Zhang Y. Artificial intelligence and biomarkers in precision nutrition. Nutr Rev. 2023;81(2):147-60.

Wang W. Fusion of clinical and imaging data improves retinal disease prediction. IEEE Access. 2021;9:11245-56.

Seddon JM. Dietary fat and risk for advanced age-related macular degeneration. Arch Ophthalmol. 2001;119(8):1191-9. DOI: https://doi.org/10.1001/archopht.119.8.1191

Kotecha A. AI-assisted multimodal retinal diagnostics. Front Digital Health. 2023;5:1221.