Concurrent Validity of Zeno Instrumented Walkway and Video-Based Gait Features in Adults With Parkinson’s Disease

Abstract

Background: Parkinson’s disease (PD) presents with motor symptoms such as bradykinesia, rigidity, and tremor that can affect gait. To monitor changes associated with disease progression or medication use, quantitative gait assessment is often performed during clinical visits. Conversely, vision-based solutions have been proposed for monitoring gait quality in non-clinical settings. Methods: We use three 2D human pose-estimation libraries (AlphaPose, Detectron, OpenPose) and one 3D library (ROMP) to calculate gait features from color video, and correlate them with those extracted by a Zeno instrumented walkway in older adults with PD. We calculate video-based gait features using a manual and automated heel-strike detection algorithm, and compare the correlations when the participants walk towards and away from the camera separately. Results: Based on analysis of 67 bidirectional walking bouts from 25 adults with PD, moderate to strong positive correlations were identified between the number of steps, cadence, as well as the mean and coefficient of variation of step width calculated from Zeno and video using 2D pose-estimation libraries. We noted that our automated heel-strike annotation method struggled to identify short steps. Conclusion: Gait features calculated from 2D joint trajectories are more strongly correlated with the Zeno than analogous gait features calculated from ROMP. Based on our analysis, videos processed with 2D pose-estimation libraries can be used for longitudinal gait monitoring in individuals with PD. Future work will seek to improve the prediction of gait features using a comprehensive machine learning model to predict gait features directly from color video without relying on intermediate extraction of joint trajectories.

Validity of AI-Based Gait Analysis for Simultaneous Measurement of Bilateral Lower Limb Kinematics Using a Single Video Camera

“…In addition, compared to traditional marker-based optical three-dimensional motion analysis or manual two-dimensional video analysis, they have great potential to solve many problems, including cost, time, efficiency, human bias, and accuracy. Appropriate use of this technology has been reported in areas such as human posture monitoring [ 8 , 9 ], rehabilitation [ 6 , 10 ], training [ 11 , 12 ], and injury prevention [ 13 ]. A well-known example of a pose estimation approach is Carnegie Mellon University’s OpenPose.…”

Evaluating the ability of a predictive vision-based machine learning model to measure changes in gait in response to medication and DBS within individuals with Parkinson’s disease

“…In contrast, during data collection for the MDC dataset, the tripod-mounted camera was positioned to the side of the instrumented gait mat on which the participants walked, but its position and orientation were not strictly controlled between walks and participants. A diagram and sample frame from this data collection methodology is available in [ 17 ], demonstrating the slight angle of the camera with respect to the direction of movement. While the lack of standardization in the camera position and orientation with respect to the direction of movement is similar to a potential real-world data collection environment, this also increased the variance in the input (as compared to the DIP dataset) and thus likely the output predictions as well.…”

A Comprehensive Review of Deep Learning Techniques for Identifying Parkinson's Disease from Gait Analysis

“…A model for Parkinson's disease prediction was proposed by Sabo et al [36] using zeno-instrumented walkways and video-based gait features in individuals with PD. This method included comparing gait characteristics calculated from color video in elderly individuals with PD using 2D human poseestimation libraries with those obtained from a Zenoinstrumented sidewalk.…”