Deep Learning-Based Estimation of Ground Reaction Forces in Parkinsonian Gait Using an Optimized Set of IMU Data

2026-08-03

AI Summary

This study addresses the limitations of traditional Parkinson’s disease gait analysis, which relies on laboratory-based equipment to measure ground reaction forces (GRF), hindering real-world monitoring. Existing inertial measurement unit (IMU)-based approaches often require multiple sensors, limiting practicality. To overcome this, the authors propose the first deep learning framework that leverages a hybrid CNN-BiLSTM model to accurately estimate bilateral vertical GRF (vGRF) from an optimized wearable IMU configuration. The research reveals distinct optimal IMU placements for patients versus healthy individuals and demonstrates that robust performance can be achieved with only 2–4 IMUs: within-subject R² reaches 0.98, while cross-subject generalization yields R² of 0.93 in healthy controls and 0.91 in patients. These results significantly enhance the feasibility and compliance of pathological gait monitoring in daily settings.

Abstract

Accurate gait analysis in Parkinson's disease (PD) typically relies on laboratory-based systems to capture biomechanical data, such as ground reaction forces (GRFs). Estimating GRFs using inertial measurement units (IMUs) provides a feasible alternative. However, this approach remains challenging in pathological gait like PD due to its high variability and complexity. Moreover, existing monitoring approaches often require multiple body-mounted sensors, which limit practicality and reduce patient compliance. To date, no study has investigated the application of deep learning approaches to address this challenge. This study proposes, for the first time, a deep learning framework to estimate bilateral vertical GRFs (vGRFs) in PD using an optimized set of wearable IMUs. A hybrid CNN-BiLSTM model was trained separately on data from 61 PD patients and 65 healthy controls (HC) using 13 IMUs. The model achieved high intra-subject accuracy ($R^2$ = 0.98) and strong inter-subject generalization ($R^2$ = 0.93 for HC, $R^2$ = 0.91 for PD). Sensor configuration was found to significantly influence estimation accuracy, with optimal sensor placement varying between PD patients and HC. For PD patients, estimation accuracy dropped markedly when reducing to a single IMU. The optimal configuration for PD used four IMUs. We identified a minimal setup with only two IMUs still enabled robust estimation. This compact setup offers a practical and scalable solution. Overall, the proposed approach supports the development of wearable vGRF-based gait analysis systems for Parkinsonian gait and potentially other pathological conditions, enabling accessible clinical assessments, remote monitoring, and personalized rehabilitation.

Problem

• Parkinson's disease
• ground reaction forces
• inertial measurement units
• pathological gait
• wearable sensors

Innovation

• deep learning
• inertial measurement units
• ground reaction forces
• Parkinson's disease
• wearable gait analysis