Back

WearGait-PD: An Open-Access Wearables Dataset for Gait in Parkinson's Disease and Age-Matched Controls

Anderson, A. J.; Eguren, D.; Gonzalez, M. A.; Khan, N.; Watkinson, S.; Caiola, M.; Hirczy, S. S.; Zabetian, C. P.; Mills, K.; Moukheiber, E.; Moro-Velazquez, L.; Dehak, N.; Motley, C.; Muir, B. C.; Butala, A. A.; Kontson, K. L.

2024-09-12 neurology
10.1101/2024.09.11.24313476 medRxiv
Show abstract

Wearable movement sensors are powerful tools for objectively characterizing and quantifying movement. They enhance the precise characterization of gait, balance, and motor symptoms in Parkinsons disease and related disorders, facilitating in-clinic and remote assessments, disease management, and therapeutic intervention development. Access to high-quality data from these sensors can accelerate discoveries in this clinical population. The WearGait-PD open-access dataset contains raw inertial measurement unit (IMU) and sensorized insole data from individuals with PD and age-matched controls, synchronized to a gait walkway reference system. IMU data include 3-degree of freedom (DOF) acceleration, rotational velocity, magnetic field strength, and orientation for each of 13 sensors on the participants body. Sensor insole data include absolute pressure from 16 sensors in each insole and 3-DOF acceleration and rotational velocity. Walkway data include 2D position and relative pressure for each active sensor during every footfall. Frame-by-frame annotation of participant actions during gait and balance tasks was incorporated using synchronized video cameras. All data were associated with demographic information and clinical evaluations (e.g., medications, DBS-status, MDS-UPDRS scores).

Published in Scientific Data (predicted rank #1) · training set

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.