Deep Learning-enabled Accurate Assessment of Gait Impairments in Parkinson's Disease

Jianda Han1,2,†, Zhihua Tian1,†, Jialing Wu3,†, Kai Zhang1, Shaohua Li1, Fahd Baig4, Peipei Liu3, Ravi Vaidyanathan5,6, Francesca Morgante4, Weiguang Huo1,2,*
1 College of Artificial Intelligence, Nankai University, Tianjin, China 2 Institute of Intelligence Technology and Robotic Systems, Shenzhen Research Institute, Nankai University, Shenzhen, China 3 Department of Neurology, Tianjin Huanhu Hospital, Tianjin, China 4 Neurosciences and Cell Biology Institute, Neuromodulation and Motor Control Section, City St George's University of London, London, UK 5 Department of Mechanical Engineering, Imperial College London, London, UK 6 Dementia Research Institute Care Research & Technology Centre, Imperial College London, London, UK

These authors contributed equally to this work

Abstract

Background: Gait impairments are among the most prevalent and disabling symptoms in Parkinson's Disease (PD), featuring complex and highly heterogeneous manifestations.

Methods: We propose a deep learning-based framework to assess gait impairments using smartphone-recorded videos. This framework demonstrated high proficiency in predicting PD severity.

Results: The model achieved a micro-average area under the receiver operating characteristic curve (AUC) of 0.87 and an F1 score of 0.806, comparable to the average performance of three clinical specialists. Additionally, it effectively discerned the comprehensive efficacy of medications on gait impairments with a precision of 73.68%.

0.87
Micro-average AUC
0.806
F1 Score
73.68%
Medication Effect Precision
Expert-level
Performance

Introduction

Parkinson's disease (PD) is the second most prevalent progressive neurodegenerative disorder, affecting more than 10 million people worldwide. Gait impairments are among the most common and disabling symptoms in PD, characterized by intricate underlying mechanisms and substantial individual variation in clinical manifestations.

Problem

Current clinical rating scales like UPDRS have low sensitivity, inherent subjectivity, and dependency on clinical specialists, limiting their utility in routine assessment of gait impairments.

Solution

Our framework uses smartphone-recorded videos and deep learning to provide objective, precise assessment of gait impairments, enabling home-based monitoring.

Innovation

We extract both traditional clinically used motion markers and discover novel digital biomarkers sensitive to disease progression and medication response.

Methodology

We developed a novel Siamese contrastive network architecture that efficiently extracts precise spatiotemporal motion characteristics of the entire body joints from gait videos recorded with a single smartphone.

Siamese Contrastive Network

Our architecture fuses gait videos recorded from both left and right lateral perspectives during shuttle walks, ensuring accurate identification of lateral motion characteristics.

Pose Estimation

We extract body skeletons from smartphone videos using DWPose, with modifications to correct misdetections of leg keypoints for accurate feature extraction.

Interpretable Framework

Our model analyzes personalized joint impacts on gait impairment severity over walking time, enabling extraction of both traditional and novel digital biomarkers.

Key Results

Severity Assessment

The model demonstrated high accuracy in predicting gait impairment severity, with performance comparable to clinical experts (F1 score: 0.806).

Medication Effect Discrimination

Our framework accurately identified medication-induced changes in gait impairments with 73.68% precision, including subtle responses undetectable by UPDRS.

Biomarker Discovery

We identified novel digital biomarkers sensitive to disease progression and medication response, such as linear velocities and accelerations of skeletal joints.

Clinical Validation

The model's performance was validated against the consensus of three clinical specialists, demonstrating alignment with expert assessments and the ability to handle inter-rater variability.

Online Assessment System

We developed an online assessment system (FAGI-PD) that enables home-based, objective, routine assessments of gait impairments in Parkinson's disease using smartphone-recorded videos.

Accessible Monitoring

Enables frequent, convenient assessment of gait impairments in home settings, reducing the need for clinical visits.

Immediate Feedback

Provides immediate assessment results and trends over time for patients and clinicians.

Longitudinal Tracking

Tracks disease progression and treatment response over time, enabling personalized therapy adjustments.

Resources

Code & Data

Code for data preprocessing, model training, and feature extraction available at Code Ocean

Access Code

Online System

Try our FAGI-PD system for gait impairment assessment

Try Demo