PERSONALIZED NUTRITION AND ARTIFICIAL INTELLIGENCE: MODERN APPROACHES, REGULATORY CHALLENGES, AND CLINICAL EFFICACY

Abstract

The digitalization of healthcare is driving a shift from generalized dietary models to personalized nutrition. Traditional approaches fail to account for individual metabolic variations, which reduces their effectiveness in chronic disease prevention. The objective of this paper is to analyze artificial intelligence systems in generating personalized dietary recommendations, evaluate their clinical efficacy, address data standardization challenges, and explore ethical implementation aspects. The study employs a systematic review of scientific sources, a comparative analysis of clinical trials—specifically randomized controlled trials—and conceptual modeling of the operational data flow. It was established that dietary personalization operates as a dynamic 5-step loop: multimodal data collection, AI-driven analysis, prediction, recommendation generation, and real-time feedback. Certain artificial intelligence models (XGBoost, Stacking) demonstrated a metabolic risk prediction accuracy of up to 96.07%. Dietary assessment utilizing computer vision and deep learning improves food recognition accuracy by 30% when accounting for national context and geolocation.Clinical data confirm that dietary recommendations developed using artificial intelligence models achieve a 39% reduction in irritable bowel syndrome symptoms and a 72.7% diabetes type 2 remission rate. Thus, the efficacy of incorporating artificial intelligence technologies into personalized recommendations for health improvement and symptom reduction in chronic diseases has been proven, with results demonstrating significant enhancements in patients' metabolic parameters and life quality.The necessity of data standardization and regulatory framework development for the successful implementation of personalized nutrition into clinical practice is also emphasized. Ethical issues regarding privacy and accountability in AI applications are highlighted, alongside the importance of explainable artificial intelligence in fostering patient trust in digital health solutions. Ultimately, the prospects for personalized nutrition as a new paradigm in dietary science with the potential to substantially improve public health and well-being approaches are outlined.Received 12.07.2026, accepted  21.08.2026, published 22.09.2026