Wearable Sensor-Based Knee Joint Angle Estimation: Modalities, Modeling, and Applications.

Area of Science:

• Biomechanics and Rehabilitation Engineering
• Wearable Sensor Technology
• Biomedical Signal Processing

Background:

• Knee Joint Angle Estimation (KAE) using wearable sensors is crucial for rehabilitation, sports biomechanics, and mobility monitoring.
• Existing research exhibits fragmentation due to diverse sensor setups, data types, and validation methods, hindering progress.
• A comprehensive synthesis of current KAE literature is needed to address these inconsistencies.

Purpose of the Study:

• To review and synthesize 128 peer-reviewed studies on wearable sensor-based KAE published between 2008 and 2024.
• To provide a consolidated perspective on KAE application domains, sensing modalities, data sources, and modeling approaches.
• To identify key challenges, evaluate common metrics, and propose recommendations for improving reproducibility and clinical translation of KAE.

Main Methods:

• Systematic literature review of 128 studies focusing on wearable sensor-based Knee Joint Angle Estimation.
• Categorization of studies based on application domains, sensing modalities (sEMG, IMUs, pressure, optical, flexible sensors), data sources, and modeling techniques (regression to deep networks).
• Consolidation and discussion of commonly used evaluation metrics (MAE, RMSE, R², NRMSE, PCC/CCC) and analysis of reproducibility challenges.

Main Results:

• The review covers diverse application domains, sensing modalities, data sources, and modeling approaches, ranging from classical regression to deep learning.
• Commonly used evaluation metrics are identified and their suitability for routine monitoring versus safety-critical applications is discussed.
• Key challenges identified include reliance on controlled datasets, limited subject-independent testing, and insufficient reporting on robustness to drift and placement variability.

Conclusions:

• Standardized multimodal benchmarks and consistent validation protocols are essential for reliable KAE.
• Addressing challenges in dataset collection, subject independence, and robustness reporting is critical for clinical translation.
• Future research should focus on developing robust, generalizable KAE systems through standardized testing and clinically relevant validation.