The real-time application and effectiveness assessment of an intelligent physical fitness testing system in physical training

Abstract

An intelligent physical fitness testing system leverages advanced technologies to monitor and evaluate individuals’ fitness levels accurately. It integrates real-time data acquisition, and analysis, to support personalized physical training and health management. This study aims to evaluate the practical application and effectiveness of an intelligent system for real-time physical fitness testing in the context of physical training. Our suggested model employs portable sensing devices and we proposed a novel Northern Goshawk optimization-driven Gate Customized Long Short-Term Memory (NG-GC-LSTM) for enhancing accuracy in evaluating the individuals’ physical fitness levels. Data acquisition involves gathering bio-sensing data from 25 individuals during diverse physical training activities. The Min-Max Scaling algorithm is utilized to pre-process the obtained sensor data. We employed a Short-Time Fourier Transform (STFT) for extracting crucial features from the processed data. In our proposed framework, the NG optimization algorithm iteratively fine-tunes the GC-LSTM architecture for the accurate evaluation of an established intelligent physical fitness testing system. The recommended model is executed in Python software. During the result analysis phase, we assess the efficacy of our model’s performance across a variety of parameters. Additionally, we conduct comparative analyses with existing methodologies. The obtained outcomes demonstrate the efficacy and superiority of the suggested framework.