FDA Wearable Gait Dataset — Synchronized Smartphone IMU & Pressure Walkway [400 Trials]
Catalog Summary
"Open-access wearable gait dataset with 400 CSV files from 20 healthy participants across 5 smartphone orientations and 2 gait trial types. 100 Hz IMU sampling. Used for smartphone-based gait metric validation and wearable sensor accuracy studies."
Catalog Notes
Overview
This open-access dataset was produced by the U.S. Food and Drug Administration (FDA) Division of Biomedical Physics to evaluate factors impacting smartphone-based gait measurement accuracy. It contains time-synchronized raw IMU data from smartphones alongside reference IMU sensors and a pressure-sensing walkway, enabling rigorous validation of wearable gait algorithms.
The study enrolled 20 healthy participants who performed normal and obstacle-avoidance gait trials while carrying smartphones in 2 different body placements and 5 different orientations. The 100 Hz sampling frequency and the inclusion of a gold-standard pressure walkway reference make this dataset especially suitable for benchmarking consumer wearable gait metrics.
The dataset is hosted on GitHub (FDA's dbp-osel organization) and linked to a PhysioNet database entry, making it fully open and reproducible. Supporting documentation covers the experimental protocol, synchronization methods, and data structure in detail.
Column Schema
| Column | Description |
|---|---|
| timestamp | Sample timestamp at 100 Hz. |
| acc_x | Smartphone accelerometer X-axis reading. |
| acc_y | Smartphone accelerometer Y-axis reading. |
| acc_z | Smartphone accelerometer Z-axis reading. |
| gyro_x | Gyroscope X-axis reading. |
| gyro_y | Gyroscope Y-axis reading. |
| gyro_z | Gyroscope Z-axis reading. |
| walkway_pressure | Reference pressure-sensing walkway measurement. |
| placement | Smartphone placement on body (2 positions). |
| orientation | Smartphone orientation label (5 orientations). |
| trial_type | Gait trial type: normal or obstacle avoidance. |
Key Statistics
- Total Files: 400 CSV files (370 retained after quality filtering)
- Participants: 20 healthy subjects
- Conditions: 2 placements × 5 orientations × 2 gait trial types
- Sampling Rate: 100 Hz (IMU and walkway synchronized)
- File Format: CSV
- Time Period: 2023
Use Cases
- Validation of smartphone-based wearable gait metric algorithms
- Evaluation of IMU sensor placement and orientation effects on step detection
- Development of robust gait analysis tools for clinical and consumer wearable devices
- Benchmarking digital health biomarkers derived from inertial sensors
Source & Attribution
Produced by the FDA Center for Devices and Radiological Health (CDRH), Division of Biomedical Physics. The dataset is openly available through the FDA's dbp-osel GitHub organization and cross-referenced with PhysioNet. Full experimental protocol and data structure documentation are included.
View Data Structure
To explore column names, data types, and sample rows, visit the official dataset page on Government.
Preview on GovernmentCite This Dataset
FDA Division of Biomedical Physics (2023). Open-Access Wearables Dataset to Evaluate Factors Impacting Accuracy of Smartphone Gait Metrics. [Dataset]. FDA CDRH / dbp-osel GitHub. https://github.com/dbp-osel/wearables-for-gait-synchronized-smartphone-imu-and-walkway-data
Source metadata: FDA CDRH / dbp-osel GitHub (2023)
Indexed by IoTDataset.com on Apr 10, 2026
Review the Source Record
Confirm the licence, version, access conditions, file format, and provenance at the source before use.