Personalized High-Intensity Interval Training Scheme Using Generative Artificial Intelligence and Intelligent Clustering Algorithms

Abstract

This work aims to explore the application of High-Intensity Interval Training (HIIT) based on intelligent clustering analysis in healthcare from the perspective of generative artificial intelligence. By integrating the Regularized Cycle-Consistent Generative Adversarial Network (RegCycleGAN) model and intelligent clustering technology, this work develops personalized healthcare solutions, addressing the lack of personalization and adaptability in traditional fitness programs. This model utilizes a large-scale, public, and fully anonymized health dataset, enabling it to generate structured training programs for individuals. Key parameters such as training duration, intensity (measured in %HRmax), and rest intervals define these programs. Crucially, this generation process is strictly constrained by exercise science principles, such as the guidelines from the American College of Sports Medicine (ACSM), to ensure the program's scientific validity and safety. Experimental results show that the model achieves an accuracy of 96.18% and an F1 score of 91.95% in generating plans that highly match individuals' physiological data. It is emphasized that these achievements constitute a computational proof-of-concept of this method rather than a direct proof of its clinical effectiveness. Its main limitation lies in the lack of prospective human intervention studies to verify the actual physiological benefits of these generated schemes. Thus, the proposed RegCycleGAN model, combined with intelligent clustering analysis, provides a new technical route for personalized healthcare management, potentially advancing the field of health management by enhancing the personalization and adaptability of medical solutions.

Ähnliche Arbeiten