Application of Additive Manufacturing and Deep Learning in Exercise State Discrimination

Area of Science:

• Sports Science
• Biomedical Engineering
• Wearable Technology

Background:

• Smart wearable devices are vital for athletic training and health management.
• Accurate detection of sports fatigue is crucial for performance optimization and injury prevention.
• Current wearable sensors face challenges in comfort, precision, and stable fatigue detection methods.

Purpose of the Study:

• To introduce a smart wearable sensing device for detecting sports fatigue states.
• To address limitations of current devices regarding comfort and accuracy.
• To establish stable methods for fatigue detection in athletes.

Main Methods:

• Utilized reverse engineering and additive manufacturing (3D printing) for device design.
• Developed a prototype incorporating a long short-term memory (LSTM) neural network.
• Analyzed bioelectrical signals to identify fatigue states and related indicators.

Main Results:

• The developed prototype successfully collected and analyzed bioelectrical signals.
• The system identified indicators associated with sports fatigue within the collected signals.
• The device demonstrated improved accuracy in classifying different states of sports fatigue.

Conclusions:

• The integration of 3D printing and deep learning offers a promising approach for comfortable and accurate fatigue detection.
• The developed wearable device and analysis method can effectively monitor sports fatigue.
• This technology has the potential to enhance athletic training, safety, and performance.