Extended Application of Inertial Measurement Units in Biomechanics: From Activity Recognition to Force Estimation

Area of Science:

• Biomechanics and Wearable Technology
• Musculoskeletal Health and Rehabilitation

Background:

• Abnormal posture and movement often signal musculoskeletal injuries or diseases.
• Mechanical forces are critical in musculoskeletal tissue injury and recovery processes.
• Machine learning models using kinematic data from Inertial Measurement Units (IMUs) show high accuracy in activity recognition and musculoskeletal force estimation.

Purpose of the Study:

• To review recent advancements in applying IMUs within biomechanics.
• To focus on IMU applications in activity recognition and musculoskeletal force estimation.
• To explore the potential of IMU-based wearable devices for musculoskeletal force prediction.

Main Methods:

• Review of methodologies including data pre-processing, noise reduction, classification, and force/torque estimation models.
• Analysis of application effects and performance of IMU-based systems.
• Examination of machine learning approaches, including adaptive networks and deep learning.

Main Results:

• IMUs have demonstrated superior accuracy in activity recognition and musculoskeletal force estimation.
• Applications span daily activity assessment, posture analysis, disease diagnosis, rehabilitation, and exoskeleton control.
• Significant technical feasibility and opportunities exist for predicting musculoskeletal forces using IMU-based wearables.

Conclusions:

• IMUs are valuable tools in biomechanics for understanding musculoskeletal health and movement.
• Accurate musculoskeletal force prediction using IMU-based wearable devices is a promising research area.
• Advancements in deep learning models will further enhance the capabilities of IMU applications in biomechanics.