Related Experiment Video
Updated: Jun 29, 2025
04:54
Iterative Development of an Innovative Smartphone-Based Dietary Assessment Tool: Traqq
Published on: March 19, 2021
4.6K
Qualitative evaluation of artificial intelligence-generated weight management diet plans
Dong Wook Kim1,2, Ji Seok Park3, Kavita Sharma2
1Division of Endocrinology, Diabetes and Hypertension, Center for Weight Management and Wellness, Brigham and Women's Hospital, Boston, MA, United States.
Frontiers in Nutrition
|April 5, 2024
Summary
Artificial intelligence (AI) can generate weight-loss diet plans comparable to those from major medical centers. While personalized AI diet plans show clinical potential, further refinement is needed for optimal applicability.
Area of Science:
- Healthcare Technology
- Artificial Intelligence in Medicine
- Nutrition Science
Background:
- Dietary interventions are crucial for weight management.
- The application of artificial intelligence (AI), especially large language models, is expanding in healthcare.
- The clinical applicability of AI-generated diet plans remains unevaluated.
Purpose of the Study:
- To assess the potential of AI-generated personalized weight-loss diet plans for clinical use.
- To compare AI-generated diet plans with those created by human experts and from tertiary medical centers.
Main Methods:
- Utilized ChatGPT (4.0) to create weight-loss diet plans.
- Selected control diet plans from tertiary medical centers for comparison.
- Invited dietitians, physicians, and nurse practitioners specializing in obesity medicine or nutrition to blindly assess plans on effectiveness, balance, comprehensiveness, flexibility, and applicability.
Main Results:
- No significant differences were found between AI-generated and human-created diet plans across evaluation categories.
- Experts could not reliably distinguish AI-generated plans from human-created ones.
- AI-generated personalized diet plans received above-neutral scores for all evaluation variables, despite noted limitations in specificity and affordability.
Conclusions:
- AI-generated diet plans show significant potential for clinical applications in weight management.
- AI diet plans were frequently indistinguishable from those used in major tertiary medical centers.
- Further research and refinement of AI inputs are necessary to enhance the quality and applicability of AI-generated dietary interventions for personalized, weight-centric care.
