FoG-STAR: Freezing of Gait Severity, Tasks, Activities, and Ratings

📌 Overview

This dataset contains wearable inertial sensor recordings and clinical/demographic information collected from 22 people with Parkinson’s disease. It is designed to support research on Freezing of Gait (FoG) detection, severity estimation, activity recognition, and digital biomarkers.

The dataset is organized in two CSV files:

• sensor_data.csv → synchronized inertial sensor signals with FoG labels and task annotations
• clinical_data.csv → subject-level demographic and clinical assessments

🎯 Key Features

• 22 participants with Parkinson's Disease
• 4 sensor positions: Left ankle, Right ankle, Back, Wrist
• 6-axis IMU data: Accelerometer (g) + Gyroscope (°/s)
• 60 Hz sampling rate
• 7 motor tasks designed to elicit FoG
• Expert annotations: FoG presence and severity levels
• Clinical assessments: H&Y, MDS-UPDRS III, FoG-Q, MoCA, FES-I, PDQ-8

📂 Dataset Structure

FoG-STAR/
├── sensor_data.csv          # Synchronized sensor signals with annotations (119.6 MB)
├── clinical_data.csv        # Subject demographics and clinical scores
├── FoG-Star_Analytics.ipynb # Example analysis notebook
└── LICENSE                  # CC-BY 4.0 license

📊 Data Description

Sensor Data (sensor_data.csv)

31 columns sampled at 60 Hz:

Column	Name	Description
1	timestamp	Timestamp in milliseconds
2-25	Sensor signals	Format: [position]_[sensor]_[axis]
26	activity	Activity code (1-7)
27	fog	Binary FoG label (0/1)
28	fog_severity	Severity during FoG (1-3)
29	subjectID	Subject identifier (1-22)
30	sessionID	Recording session ID
31	taskID	Task code (1-7)

Sensor naming convention: [position]_[sensor]_[axis]

• Positions: ankleL, ankleR, back, wrist
• Sensors: acc (accelerometer), gyro (gyroscope)
• Axes: x, y, z

Activity codes:

1. Walking
2. Sitting
3. Standing
4. Sit-to-Stand transition
5. Stand-to-Sit transition
6. Right turn
7. Left turn

FoG severity levels:

1. Shuffling forward
2. Trembling in place
3. Complete akinesia

Task codes:

1. Timed Up-and-Go (TUG)
2. Standing for 1 minute
3. Walking back and forth
4. Walking through doorway
5. Walking while carrying water
6. Walking while counting backwards
7. 360° turn

Clinical Data (clinical_data.csv)

10 variables for 22 subjects:

Column	Variable	Description	Range
1	subjectID	Subject identifier	1-22
2	age	Age in years	-
3	gender	Gender	M/F
4	disease_duration	Years since PD diagnosis	-
5	h_y	Hoehn & Yahr stage	0-5
6	updrs_iii	MDS-UPDRS Part III score	0-76
7	fog_q	Freezing of Gait Questionnaire	0-24
8	moca	Montreal Cognitive Assessment	0-30
9	fes_i	Falls Efficacy Scale–International	16-64
10	pdq_8	Parkinson's Disease Questionnaire–8	0-32

📈 Analysis Examples

The included FoG-Star_Analytics.ipynb notebook provides:

• Data exploration and visualization
• FoG event duration analysis
• Severity distribution analysis
• Clinical correlation studies
• Signal visualization with FoG annotations

🔬 Research Applications

This dataset supports various research directions, such as:

1. FoG Detection: Binary classification of FoG presence
2. Severity Estimation: Multi-class severity prediction
3. Activity Recognition: Classification of motor activities
4. Predictive Modeling: FoG prediction before onset
5. Digital Biomarkers: Correlation with clinical assessments
6. Multi-modal Fusion: Combining multiple sensor positions

📝 Citation

If you use this dataset in your research, please cite:

@dataset{borzi2025fogstar,
  author       = {Borzì, Luigi and Demrozi, Florenc and others},
  title        = {FoG-STAR: Freezing of Gait Severity, Tasks, Activities, and Ratings},
  year         = {2025},
  publisher    = {Zenodo},
  doi          = {10.5281/zenodo.16989602},
  url          = {https://doi.org/10.5281/zenodo.16989602}
}

@article{borzi2025freezing,
  title={Freezing of gait detection: The effect of sensor type, position, activities, datasets, and machine learning model},
  author={Borzì, Luigi and others},
  journal={Journal of Parkinson's Disease},
  volume={15},
  number={1},
  pages={163--181},
  year={2025}
}

@article{demrozi2023lowcost,
  title={A low-cost wireless body area network for human activity recognition in healthy life and medical applications},
  author={Demrozi, Florenc and others},
  journal={IEEE Transactions on Emerging Topics in Computing},
  volume={11},
  number={4},
  pages={839--850},
  year={2023}
}

⚖️ License

This dataset is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license.

You are free to:

• Share: Copy and redistribute the material in any medium or format
• Adapt: Remix, transform, and build upon the material for any purpose

Under the following terms:

• Attribution: You must give appropriate credit and provide a link to the license

👥 Contributors

• Luigi Borzì¹
• Florenc Demrozi²
• Ruggero Bacchin³
• Cristian Turetta⁴
• Michele Tebaldi⁴
• Luis Sigcha⁵
• Samaneh Zolfagharian⁶
• Domiziana Rinaldi⁷
• Giuliana Fazzina⁸
• Giulio Balestro⁹
• Alessandro Picelli⁹
• Graziano Pravadelli⁴,¹⁰
• Gabriella Olmo¹¹
• Stefano Tamburin⁹
• Leonardo Lopiano¹²
• Carlo Alberto Artusi¹³

¹Politecnico di Torino Dipartimento di Automatica e Informatica, ²University of Stavanger, ³Ospedale Santa Chiara, ⁴University of Verona, ⁵University of Limerick Faculty of Education and Health Sciences, ⁶School of Innovation Design and Engineering, Malardalen University, Vaster as, Sweden, ⁷Department of Neuroscience, Mental Health and Sensory Organs, Sapienza University of Rome, ⁸Department of Neuroscience, University of Turin, ⁹Department of Neurosciences, Biomedicine and Movement Sciences, University of Verona, ¹⁰EDALab s.r.l., ¹¹Department of Control and Computer Engineering, Politecnico di Torino, ¹²University of Turin, ¹³Department of Neuroscience, University of Turin

📧 Contact

For questions about the dataset, please:

• Open an issue on this repository
• Contact the corresponding authors through the Zenodo record
• Visit the Zenodo dataset page

🔗 Links

• Zenodo Dataset
• Related Publications
• IEEE TETC Paper