Classification of Upper Limb Movements Using Convolutional Neural Network with 3D Inception Block

Abstract

A brain-machine interface (BMI) based on electroencephalography (EEG) can overcome the movement deficits for patients and real-world applications for healthy people. Ideally, the BMI system detects user movement intentions transforms them into a control signal for a robotic arm movement. In this study, we made progress toward user intention decoding and successfully classified six different reaching movements of the right arm in the movement execution (ME). Notably, we designed an experimental environment using robotic arm movement and proposed a convolutional neural network architecture (CNN) with inception block for robust classify executed movements of the same limb. As a result, we confirmed the classification accuracies of six different directions show 0.45 for the executed session. The results proved that the proposed architecture has approximately 6∼13% performance increase compared to its conventional classification models. Hence, we demonstrate the 3D inception CNN architecture to contribute to the continuous decoding of ME.

EEG Identity Authentication in Multi-Domain Features: A Multi-Scale 3D-CNN Approach

“…Inadequate advanced studies have found that the performance of parallel multi-scale convolution structure was better than serial mode, and the parallel structure of multi-scale is beneficial to capture the characteristics of different scales. The inception module (Lee et al, 2020 ) in the field of image was used in the task of brain control manipulator to obtain better stability and accuracy in decoding motion intention. In addition, Zhao et al ( 2019b ) research designed small receptive field network (SRF), medium receptive field network (MRF), and large receptive field network (LRF), using multi-scale three-dimensional dimensions (2 × 2 × 1, 2 × 2 × 3, 2 × 2 × 5) to extract EEG features.…”

Enhancing ERP-based Brain-Computer Interfaces for Practical Applications: Asynchrony, Deep Learning, and a Novel BCI Platform

“…The Inception architecture was originally introduced by Szegedy et al (2015) for image processing, resulting in significant advancements in that field. However, before our study there were only a few studies that have used this architecture for EEG processing, and none for ERP detection (Lee et al, 2020;Qiao and Bi, 2019;Yue and Wang, 2019). EEG-Inception includes two Inception modules with three branches that process the signal at different time scales (i.e.…”