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.

Keywords: Computer vision; Parkinson’s disease; Zeno instrumented walkway; human pose-estimation; quantitative gait analysis.

FIGURE 1.

The top panel presents an overhead view of data collection environment. Participants were instructed to walk along a 6 m instrumented walkway while simultaneously being recorded by a standard color camera. The bottom panels show the major joints detected by the OpenPose pose-estimation library as a participant walks toward (left) and away from (right) the camera.

FIGURE 2.

Horizontal ankle positions for a sample walk of a participant walking away from the camera. The stances detected by the ST-DBSCAN algorithm are denoted by red boxes, and heel-strikes were selected as the first timestep of each detected stance. Note that as the participant walks farther from the camera (after ~5 seconds), the resolution of the joint trajectory signal is insufficient for the ST-DBSCAN algorithm to identify the last step (instead it is combined into the previous stance).

FIGURE 3.

Scatterplots of number of steps detected by Zeno and video analysis, grouped by heel-strike annotation method. The red line represents the fit and confidence bounds of the linear regression model between the two variables.