A High-Precision Wearable System for Gait Analysis Integrating High-Spatial-Resolution Pressure Insoles With Embedded Walking-Pattern-Informed Algorithms

Contemporary wearable devices for gait analysis still encounter precision limitations that preclude their qualification as medical practicing tools. This study presents a fidelity-tracking wearable system (FTW) for gait analysis with novel walking-pattern-informed algorithms leveraging pressure insoles integrated with inertial measurement unit (IMU). Customized pressure insoles achieved high spatial resolution (5.2 points/cm²) based on an interleaved network multiplexing electrode (INME) sensor design. Controlled experiments involving nine healthy subjects demonstrated that this design significantly improves the estimation accuracy of both pressure-related and temporal gait parameters. Walking-pattern-informed algorithms were developed to further enhance the accuracy of parameter estimation. Specifically, pressure and area thresholds were respectively used for detecting toe-off and strike moments, optimizing the accuracy of temporal parameter calculations. A gradually zeroing the pressure in the metatarsal region (GZPM) strategy was employed to mitigate bending noise at peak thrust, enhancing the accuracy of pressure parameter calculations. Additionally, IMU-determined toe-off timing was advanced by 25% to better align with typical gait characteristics and optimize step width calculations. A classification algorithm was developed based on four fundamental gait conditions to calculate clinically defined foot progression angles. FTW enables the simultaneous and precise output of 13 gait-related parameters. Compared to previous studies, errors in step time and stance phase were reduced by 49.4% and 36.2%, respectively. The root mean squared (RMS) error of the measured plantar pressure center trajectory length was 0.48 ± 0.48 cm, and the accuracy of the spatial parameters was consistent with previous studies. This high-precision, multi-parameter wearable gait monitoring system holds significant promise for home-based long-term monitoring of chronic musculoskeletal and neurological disorders.