Evaluation of the compliance of diet plans and nutritional advice generated by artificial intelligence with guidelines for cardiac patients

Abstract

Introduction In response to the increasing demand for personalization in cardiac dietetics, tools based on artificial intelligence (AI) are gaining popularity. We assess the compliance of AI-generated nutritional advice and diet plans with current dietary standards. Objective The study aims to evaluate whether nutritional advice and diet plans created by AI for heart disease patients align with the recommended caloric intake and macronutrient ratios, based on current dietary guidelines. Methods Using selected AI tools, nutritional advice and diet plans were generated for defined dietary scenarios corresponding to specific cardiac diagnoses. The analysis included assessing the caloric content and macronutrient composition (protein, fat, carbohydrates) compared to current guidelines. Results The analysis showed that the majority of the AI-generated diet plans and nutritional advice were compliant with dietary recommendations, with an >75% compliance rate for caloric content and macronutrient ratios. However, instances of non-compliance were discovered, suggesting a need for further adjustment of AI algorithms. Conclusions AI tools demonstrate potential in creating personalized diet plans for cardiac patients but require optimization to fully meet current dietary guidelines. The study highlights the importance of integrating dietary knowledge into the design process of AI tools, which is crucial for providing high-quality nutritional support.