Gait Data for Parkinson's Detection Using FMCW and CW Doppler Radar Networks
by Lopez Delgado, Ignacio Esteban / e-cienciaDatos Harvested Dataverse·Updated 2mo ago
Available on 1 platform
Sign in to view source links and access this dataset
Description
A dataset from Universidad Politécnica de Madrid, captured to compare Frequency-Modulated Continuous-Wave (FMCW) and Continuous-Wave (CW) radar networks for gait analysis. It contains approximately 100 minutes of gait recordings from ten participants, split evenly between five healthy individuals and five Parkinson's disease patients, with Vicon infrared cameras providing ground truth. The dataset, authored by Lopez Delgado, Ignacio Esteban, includes raw signals and over 2000 gait parameter estimates.
Use Cases
Compare the efficacy of FMCW versus CW Doppler radar for gait feature extraction based on the dual radar network setup.
Develop machine learning models for Parkinson's disease detection based on radar-captured gait parameters.
Validate radar-based gait analysis systems against ground-truth Vicon motion capture data.
Study age and health-related gait differences based on the participant groups (healthy young vs. older Parkinson's patients).
Strengths
Includes ground-truth validation from a six-camera Vicon infrared motion capture system recording at 120 Hz.
Contains data from ten participants, with five healthy and five Parkinson's disease patients, providing a comparative basis.
Offers both raw radar signals and over 2000 pre-processed gait parameter estimates.
Uses two distinct 24-GHz radar network types (FMCW with 125 MHz bandwidth and CW) configured to avoid interference.
Limitations
Column-level documentation is absent; field semantics must be inferred after download.
Row count is unknown, which may limit suitability assessment.
Data access is currently restricted, requiring a request form and password, with free availability only from 2027.
Gait data captured using synchronized Vicon infrared cameras and two types of 24-GHz radar networks during five clinical tests.
Time Range
null
Freshness
Last updated 2026-04 21 17:47:22; freshness should be verified.
Geography
null
Data files are password-protected; access requires filling and submitting a 'data_request_form' to the specified email. Data will be freely available only from 2027 onward.