EEG-based trial-by-trial texture classification during active touch

Safaa Eldeeb,Murat Akcakaya,Andac Demir,Jordyn Ting,Deniz Erdogmus,Douglas Weber

doi:10.1038/s41598-020-77439-7

Safaa Eldeeb, Murat Akcakaya + Show 4 more

Open Access

https://doi.org/10.1038/s41598-020-77439-7

Copy DOI

Abstract

Trial-by-trial texture classification analysis and identifying salient texture related EEG features during active touch that are minimally influenced by movement type and frequency conditions are the main contributions of this work. A total of twelve healthy subjects were recruited. Each subject was instructed to use the fingertip of their dominant hand’s index finger to rub or tap three textured surfaces (smooth flat, medium rough, and rough) with three levels of movement frequency (approximately 2, 1 and 0.5 Hz). EEG and force data were collected synchronously during each touch condition. A systematic feature selection process was performed to select temporal and spectral EEG features that contribute to texture classification but have low contribution towards movement type and frequency classification. A tenfold cross validation was used to train two 3-class (each for texture and movement frequency classification) and a 2-class (movement type) Support Vector Machine classifiers. Our results showed that the total power in the mu (8–15 Hz) and beta (16–30 Hz) frequency bands showed high accuracy in discriminating among textures with different levels of roughness (average accuracy > 84%) but lower contribution towards movement type (average accuracy < 65%) and frequency (average accuracy < 58%) classification.

Highlights

Trial-by-trial texture classification analysis and identifying salient texture related EEG features during active touch that are minimally influenced by movement type and frequency conditions are the main contributions of this work
The results of the systematic feature selection, which is explained in details under the classification subsection of the “Materials and methods” section, showed that the spectral EEG features, (the total power in the mu (8–15 Hz) and beta (16–30 Hz) frequency bands), showed high accuracy in discriminating among textures with different levels of roughness while movement type and frequency classification accuracies were low, see Fig. 1
Since the minimum classification accuracies for movement type and frequency classifications are obtained in groups 1 and 2, we here focus on the results obtained in these two groups

Summary

Introduction

Trial-by-trial texture classification analysis and identifying salient texture related EEG features during active touch that are minimally influenced by movement type and frequency conditions are the main contributions of this work. During active stimulation of the skin, both slowly and rapidly adapting mechanoreceptors contribute to the touch sensation process. Blatow et al.[17], performed a tactile stimulation study on the fingers of healthy subjects during tactile stimulation of both right and left fingers They showed higher activation in the contralateral side of the somatosensory cortex during stimulation. A device that provides a dynamic passive stimulation that mimics the movement of sliding an object against a participant’s finger while recording EEG was introduced[18]

Objectives

Methods

Results

Discussion

Conclusion