Boosting Automatic Exercise Evaluation Through Musculoskeletal Simulation-Based IMU Data Augmentation

TL;DR

This paper introduces a musculoskeletal simulation-based data augmentation technique for IMU data that enhances deep learning models' ability to evaluate movement quality in physiotherapy and sports training.

Contribution

The study presents a novel, biomechanically plausible data augmentation method that improves classification accuracy and generalization in IMU-based movement assessment.

Findings

Augmented data closely resembles real-world IMU data.

Significant improvement in neural network classification accuracy.

Enhanced model generalization and patient-specific fine-tuning.

Abstract

Automated evaluation of movement quality holds significant potential for enhancing physiotherapeutic treatments and sports training by providing objective, real-time feedback. However, the effectiveness of deep learning models in assessing movements captured by inertial measurement units (IMUs) is often hampered by limited data availability, class imbalance, and label ambiguity. In this work, we present a novel data augmentation method that generates realistic IMU data using musculoskeletal simulations integrated with systematic modifications of movement trajectories. Crucially, our approach ensures biomechanical plausibility and allows for automatic, reliable labeling by combining inverse kinematic parameters with a knowledge-based evaluation strategy. Extensive evaluations demonstrate that augmented variants closely resembles real-world data, significantly improving the classification accuracy and generalization capability of neural network models. Additionally, we highlight the benefits of augmented data for patient-specific fine-tuning scenarios, particularly when only limited subject-specific training examples are available. Our findings underline the practicality and efficacy of this augmentation method in overcoming common challenges faced by deep learning applications in physiotherapeutic exercise evaluation.