A Personalized Energy Expenditure Estimation Method Using Modified MET and Heart Rate-Based DQN

Area of Science:

• Biomedical Engineering
• Wearable Technology
• Personalized Health

Background:

• Wearable devices integrate bio-signals for personalized services.
• Accurate real-time energy expenditure estimation is challenging due to movement and measurement limitations.

Purpose of the Study:

• To propose a novel real-time and personalized energy expenditure estimation (EEE) method.
• To address the limitations in accurately and rapidly estimating energy expenditure (EE) from wearable devices.

Main Methods:

• Developed Real-Time Energy Expenditure (RTEE), a novel method for EEE.
• Integrated a Deep Q-Network (DQN)-based activity intensity coefficient inference network.
• Utilized a modified energy consumption prediction algorithm with real-time heart rate variations.

Main Results:

• The RTEE method enables real-time and personalized energy expenditure estimation.
• The approach leverages DQN for activity intensity and heart rate data for prediction.
• The algorithm demonstrates applicability to various heart rate-based EEE methods.

Conclusions:

• The proposed RTEE method offers a significant advancement in wearable-based energy expenditure tracking.
• This personalized approach enhances the accuracy and responsiveness of activity monitoring.
• The integration of advanced AI with bio-signal data opens new possibilities for health and fitness applications.