The latest advancements in the field of deep learning and biomedical engineering have allowed for the development of myoelectric interfaces based on deep neural networks. A longstanding problem of these interfaces is that the models cannot easily be applied to new users due to the high variability and stochastic nature of the electromyography signals. Further training a new model for every new subject requires the collection of large volumes of data. Therefore, this work proposes a transfer learning (TL) scheme which allows reusing the knowledge of a pre-existing model for a new user. Firstly, a convolutional neural network (CNN) is trained on an initial dataset using the data of multiple subjects. Then, the weights of this model are fine-tuned for a new target subject. The approach is evaluated on the Ninapro datasets DB2 and DB7. The experimentation included three different CNN models and eight preprocessing alternatives. The results showed that the success of the TL method depends on how the data are preprocessed. Specifically, the biggest accuracy improvement (+5.14%) is achieved when only the first 20% of the signal duration is used.
Tsinganos, P, Cornelis, JPH, Cornelis, B, Jansen, B & Skodras, A 2021, Transfer Learning in sEMG-based Gesture Recognition. in 2021 12th International Conference on Information, Intelligence, Systems & Applications (IISA)., 9555555, IISA 2021 - 12th International Conference on Information, Intelligence, Systems and Applications, IEEE, 12th International Conference on Information, Intelligence, Systems & Applications (IISA), Chania, Greece, 12/07/21. https://doi.org/10.1109/IISA52424.2021.9555555
Tsinganos, P., Cornelis, J. P. H., Cornelis, B., Jansen, B., & Skodras, A. (2021). Transfer Learning in sEMG-based Gesture Recognition. In 2021 12th International Conference on Information, Intelligence, Systems & Applications (IISA) Article 9555555 (IISA 2021 - 12th International Conference on Information, Intelligence, Systems and Applications). IEEE. https://doi.org/10.1109/IISA52424.2021.9555555
@inproceedings{1db96d275f554bab987731ddba9a097e,
title = "Transfer Learning in sEMG-based Gesture Recognition",
abstract = "The latest advancements in the field of deep learning and biomedical engineering have allowed for the development of myoelectric interfaces based on deep neural networks. A longstanding problem of these interfaces is that the models cannot easily be applied to new users due to the high variability and stochastic nature of the electromyography signals. Further training a new model for every new subject requires the collection of large volumes of data. Therefore, this work proposes a transfer learning (TL) scheme which allows reusing the knowledge of a pre-existing model for a new user. Firstly, a convolutional neural network (CNN) is trained on an initial dataset using the data of multiple subjects. Then, the weights of this model are fine-tuned for a new target subject. The approach is evaluated on the Ninapro datasets DB2 and DB7. The experimentation included three different CNN models and eight preprocessing alternatives. The results showed that the success of the TL method depends on how the data are preprocessed. Specifically, the biggest accuracy improvement (+5.14%) is achieved when only the first 20% of the signal duration is used.",
keywords = "gesture recognition, electromyography, convolutional neural network, transfer learning",
author = "Panagiotis Tsinganos and Cornelis, {Jan Paul Herman} and Bruno Cornelis and Bart Jansen and Athanassios Skodras",
year = "2021",
doi = "10.1109/IISA52424.2021.9555555",
language = "English",
isbn = "978-1-6654-0033-6",
series = "IISA 2021 - 12th International Conference on Information, Intelligence, Systems and Applications",
publisher = "IEEE",
booktitle = "2021 12th International Conference on Information, Intelligence, Systems & Applications (IISA)",
note = "12th International Conference on Information, Intelligence, Systems & Applications (IISA) ; Conference date: 12-07-2021 Through 14-07-2021",
}