Unmet Need
Human gait analysis is now common in many fields of clinical and basic research, but gold standard approaches – e.g., three-dimensional motion capture, instrumented mats or footwear, and wearables – are often expensive, immobile, data-limited, and/or require specialized equipment or expertise for operation, furthermore limiting the access to gait analysis usage. Additionally, there are markerless motion capture approaches for gait analysis currently available, but these are time consuming, not intuitive, and still require additional equipment. Ultimately, a current human gait analysis is restrictive, time consuming and costly. While required for many diagnoses and monitoring of disease symptoms, such as in Parkinson’s disease, it is critical for human gait analysis to become more accessible as a diagnostic and symptom monitoring tool and further the health of patients.
Technology Overview
Researchers at Johns Hopkins University were able to utilize pose estimation, through OpenPose, a freely available human pose estimation algorithm, to accurately estimate human gait parameters utilizing two-dimensional digital video input. The software workflow enables quantitative, remote assessment of human movement using only simple videos recorded from household devices (e.g., smartphones, tablets, laptop computers) with minimal costs of time, money, or effort. Additionally, the workflow approach is easy to use, fast, and accurate when compared to golden standards of three-dimensional motion capture. The end result is the ability for physicians to conduct more frequent, quantitative assessments of human motor function in any setting (e.g., directly in the home or clinic) without the requirement of specialized equipment or expertise operation.
Currently, the inventors have focused on software development aimed at motor assessments for persons with Parkinson’s disease or stroke. These individuals require motor assessments for symptom management.
Stage of Development
The team has successfully produced functional gait and repetitive movement, which have been validated against ground-truth measurements of human movement in healthy young adults. The next step in testing is external validation within patient populations.
Publication
Jan Stenum, Cristina Rossi, Ryan T. Roemmich (2020) Two-dimensional Video-based Analysis of Human Gait Using Pose Estimation, bioRxiv, https://doi.org/10.1101/2020.07.24.218776
