Skip to content

mHealth-Research-Group/paaws-study

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

15 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

PAAWS Study Resources, Code, and Issue Tracker

This repository serves as a central resource of all things related to the Physical Activity Assessment Using Wearable Sensors (PAAWS) dataset. If you use any resources discussed in this repository, please cite our paper and email us. For more information about the PAAWS study, please see the PAAWS website.

Repository Structure

/misc_scripts: folder containing sample scripts to load raw accelerometer data into a DataFrame with timestamps and activity labels (read_accelerometer_data.py) and to rotate ankle and waist sensor data so the data is in the same orientation across the SimFL+Lab and FL protocols (rotate_sensors.py).

/qc_scripts: folder containing scripts run on our annotations as part on our quality control process.

/papers: folder containing a running list of papers our group has published about collecting, annotating, or using the PAAWS dataset with accompanying PDFs.

Data Download

A portion of the PAAWS dataset, the R1 release, is available for download. The R1 release includes data collected 252 participants SimFL+Lab protocol (~808GB), 20 participants FL protocol (~111GB), and 15 participants Sleep protocol (~22GB).

More information about the PAAWS R1 dataset can be found in our recent IMWUT '25 publication and the supplemental material.

Future Releases

We anticipate additional FL and Sleep data will be available soon. We expect the entire dataset to be available to the public in 2026. This repo will be updated each time we make a new release.

Software

Annotation Software

The PAAWS data was annotated after-the-fact by human annotator, a laborious and time-consuming task. We developed a custom annotation software to help expedite our annotation process. Our annotation software is open source and available to use.

More about our annotation software can be found in our recent ARDUOUS '25 publication.

Signaligner

We developed Signaligner-Pro, an interactive tool for algorithm-assisted exploration and annotation of raw accelerometer data.

Benchmarking the PAAWS R1 dataset

To accompany the release of the PAAWS R1 dataset, we provide trained human activity recognition models for future researchers to use.

More about benchmarking the PAAWS R1 dataset can be found in the projects GitHub repository: github.com/mHealth-Research-Group/paaws-benchmarking.

Questions or Issues

Please submit an issue in our issue tracker. If we have not gotten back to you within a week, email us about the issue directly.

Citation

If you use any portion of the PAAWS dataset, please cite the following paper:

Veronika Potter, Hoan Tran, Daniel Mobley, Suzanne M. Bertisch, Dinesh John, and Stephen Intille. 2025. The Physical Activity Assessment Using Wearable Sensors (PAAWS) Dataset: Labeled Laboratory and Free-Living Accelerometer Data. Proc. ACM Interact. Mob. Wearable Ubiquitous Technol. 9, 4, Article 204 (December 2025), 32 pages. https://doi.org/10.1145/3770639

Or, as bibtex:

@article{potter_2025_paaws_dataset,
  title = {The Physical Activity Assessment Using Wearable Sensors (PAAWS) Dataset: Labeled Laboratory and Free-Living Accelerometer Data},
  author = {Potter, Veronika, and Tran, Hoan and Mobley, Daniel and Bertisch, Suzanne M. and John, Dinesh and Intille, Stephen},
  year = {2025},
  month = Dec,
  journal = {Proc. ACM Interact. Mob. Wearable Ubiquitous Technol.},
  volume = {9},
  number = {4},
  doi = {10.1145/3770639}
}

Stay up to Date!

If you'd like to remain up-to-date about new releases and work using the dataset join our mailing list.

Acknowledgements

The PAAWS dataset was supported by the National Cancer Institute of the National Institutes of Health under award number R01CA252966. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

About

A central hub for everything related to the the Physical Activity Assessment Using Wearable Sensors (PAAWS) dataset.

Resources

License

Contributing

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages