UniBo-INAIL sEMG dataset

Public sEMG dataset [1] for research, realized in a collaboration between

• Energy-Efficient Embedded Systems (EEES) Laboratory (IT) of University of Bologna - Prof. Luca Benini
• INAIL Prosthesis Centre in Vigorso di Budrio (EN|IT) (Bologna) - contact Prof. Emanuele Gruppioni

Structure

The UniBo-INAIL dataset has a nested structure:

• $7$ subjects (healthy males aged $29.5 \pm 12.2$ years)
  • $8$ sessions per subject, on different acquisition days
    • $4$ arm postures per session: proximal (the sole with the arm not fully extended; the most common in literature), distal, distal with palm down, distal with arm $45°$ up.

The sessions are $224$ in total, and each one is a small but complete dataset individually suitable for small Machine Learning experiments. There are $6$ classes: rest, power grip, 2-finger pinch grip, 3-finger pinch grip, pointing index, and open hand. Each movement is repeated $9$ to $16$ times.

The sEMG acquired via $4$ Ottobock 13E200 MyoBock Electrodes placed on the forearm's skin above the muscles involved in the chosen gestures: extensor carpi ulnaris, extensor communis digitorum, flexor carpi radialis, and flexor carpi ulnaris. The sampling frequency is $500 \text{Hz}$.

Other works and documentation on the UniBo-INAIL dataset:

• first paper on the dataset, by B. Milosevic et al. [2];
• M.Sc. thesis based on the dataset, by M. Zanghieri [3];
• papers with earlier versions of the UniBo-INAIL acquisition setup and protocol, by S. Benatti et al. [4], [5].

Usage

The data/ folder contains a .mat file for each subject, day and arm posture.

The scripts/ folder provides Python and MATLAB functions for loading the data.

Citation

If you use this dataset, please cite our paper [1]:

@INPROCEEDINGS{zanghieri2023online,
  author={Zanghieri, Marcello and Orlandi, Mattia and Donati, Elisa and Gruppioni, Emanuele and Benini, Luca and Benatti, Simone},
  booktitle={2023 IEEE Biomedical Circuits and Systems Conference (BioCAS)}, 
  title={Online Unsupervised Arm Posture Adaptation for {sEMG}-based Gesture Recognition on a Parallel Ultra-Low-Power Microcontroller}, 
  year={2023},
  volume={},
  number={},
  pages={1-5},
  doi={10.1109/BioCAS58349.2023.10388902}}

References

[1] M. Zanghieri, M. Orlandi, E. Donati, E. Gruppioni, L. Benini, S. Benatti, “Online unsupervised arm posture adaptation for sEMG-based gesture recognition on a parallel ultra-low-power microcontroller,” in 2023 IEEE International Conference on Biomedical Circuits and Systems (BioCAS), 2023, pp. 1-5. DOI: 10.1109/BioCAS58349.2023.10388902.

[2] B. Milosevic, E. Farella, S. Benatti, “Exploring arm posture and temporal variability in myoelectric hand gesture recognition,” in 2018 7th IEEE International Conference on Biomedical Robotics and Biomechatronics (BioRob), 2018, pp. 1032–1037. DOI: 10.1109/BIOROB.2018.8487838.

[3] M. Zanghieri, “sEMG-based hand gesture recognition with deep learning,” M.Sc. thesis, University of Bologna, Bologna, Italy, 2019. DOI: 10.48550/arXiv.2306.10954.

[4] S. Benatti, B. Milosevic, E. Farella, E. Gruppioni, L. Benini, “A prosthetic hand body area controller based on efficient pattern recognition control strategies,” in Sensors, vol. 17, no. 4, art. num. 869, 2017. DOI: 10.3390/s17040869.

[5] S. Benatti, E. Farella, E. Gruppioni, L. Benini, “Analysis of robust implementation of an EMG pattern recognition based control,” in Proceedings of the International Joint Conference on Biomedical Engineering Systems and Technologies - Volume 4 2014, pp. 45–54. DOI: 10.5220/0004800300450054.

License

All files are released under the LGPL-2.1 license (LGPL-2.1) (see LICENSE).