Subject-transfer framework with unlabeled data based on multiple distance measures for surface electromyogram pattern recognition. (April 2022)

• Record Type: Journal Article
• Authors: Kanoga, Suguru; Hoshino, Takayuki; Asoh, Hideki
• Is Part Of: Biomedical signal processing and control. Volume 74(2022)
• Publication Date: 2022-04
• Subjects: Transfer learning -- Surface electromyogram (sEMG) -- Wearable sensor -- Multiple distance measures -- Pattern recognition; Signal processing -- Periodicals; Biomedical engineering -- Periodicals; Signal Processing, Computer-Assisted -- Periodicals; Image Processing, Computer-Assisted -- Periodicals; Biomedical Engineering -- Periodicals; 610.28
• DOI: 10.1016/j.bspc.2022.103522
• Languages: English
• ISSNs: 1746-8094

Abstract

Highlights: An unlabeled subject-to-subject transfer framework for surface EMG is proposed. The performance of the model using two databases are validated. The classifiers are trained with time-domain and autoregressive features. The proposed method exhibits superior performance over randomly selected subjects. The sample codes used are publicly available. Abstract: To improve the initial accuracy of wearable sensor-driven human interfaces, inter-subject variabilities must be reduced through transfer learning. If subject transfer can be performed without labeling the target user's calibration data, an interface that provides stable accuracy can be easily achieved without a cumbersome calibration protocol. Herein, we propose a subject-transfer framework based on multiple distance measures that enables subject transfer using only unlabeled calibration data by minimizing the distance between the data distributions of the target and the source. To assess the performance of this framework, we used two surface electromyogram databases (one private database and one public database called the NinaPro database 5) acquired from the same wearable sensor, the Myo Gesture Control Armband. The proposed framework improved the pattern recognition accuracy compared with well-established classifiers constructed from randomly selected source subject data. In the future, we will apply this framework to online human interfaces that are not based on a specific calibration protocol. The scripts used in this study can be downloaded from https://github.com/aistairc/Unlabeled_STM .