Related Experiment Video
Updated: Jan 15, 2026
11:54
Real-Time Proxy-Control of Re-Parameterized Peripheral Signals using a Close-Loop Interface
Published on: May 8, 2021
5.1K
Human motion heart rate measurement with MSSF and feature interaction
1College of Sport, Shaoxing University, Shaoxing, China.
Computer Methods in Biomechanics and Biomedical Engineering
|October 11, 2025
Summary
This study introduces a new heart rate monitoring model for dynamic exercise, achieving 98.7% accuracy. The advanced method enhances precision and personalization during complex motion, improving user satisfaction.
Area of Science:
- Biomedical Engineering
- Wearable Technology
- Exercise Physiology
Background:
- Conventional heart rate monitoring faces accuracy and personalization challenges during dynamic exercise.
- Existing methods struggle with complex motion and individual physiological variations.
Purpose of the Study:
- To develop a novel, accurate, and personalized heart rate monitoring model for dynamic exercise settings.
- To improve heart rate monitoring precision and adaptability in real-world, complex motion scenarios.
Main Methods:
- Utilized Multi-Scale Spectral Feature selection.
- Employed an ensemble model combining Extremely Randomized Trees and a Multilayer Perceptron.
- Applied transfer learning to personalize the model for individual physiological differences.
Main Results:
- Achieved 98.7% accuracy in real-world dynamic exercise applications.
- Demonstrated significant improvements in precision and personalization of heart rate monitoring.
- Reported high user satisfaction with the proposed monitoring approach.
Conclusions:
- The novel model offers a robust solution for accurate and personalized heart rate monitoring during complex motion.
- This approach overcomes limitations of conventional methods in dynamic exercise environments.
- Transfer learning enhances the adaptability and effectiveness of the monitoring system.
