English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Objective: Electroencephalography (EEG) and electromyography (EMG) are twonon-invasive bio-signals, which are widely used in human machine interface(HMI) technologies (EEG-HMI and EMG-HMI paradigm) for the rehabilitation ofphysically disabled people. Successful decoding of EEG and EMG signals intorespective control command is a pivotal step in the rehabilitation process.Recently, several Convolutional neural networks (CNNs) based architectures areproposed that directly map the raw time-series signal into decision space andthe process of meaningful features extraction and classification are performedsimultaneously. However, these networks are tailored to the learn the expectedcharacteristics of the given bio-signal and are limited to single paradigm. Inthis work, we addressed the question that can we build a single architecturewhich is able to learn distinct features from different HMI paradigms and stillsuccessfully classify them. Approach: In this work, we introduce a singlehybrid model called ConTraNet, which is based on CNN and Transformerarchitectures that is equally useful for EEG-HMI and EMG-HMI paradigms.ConTraNet uses CNN block to introduce inductive bias in the model and learnlocal dependencies, whereas the Transformer block uses the self-attentionmechanism to learn the long-range dependencies in the signal, which are crucialfor the classification of EEG and EMG signals. Main results: We evaluated andcompared the ConTraNet with state-of-the-art methods on three publiclyavailable datasets which belong to EEG-HMI and EMG-HMI paradigms. ConTraNetoutperformed its counterparts in all the different category tasks (2-class,3-class, 4-class, and 10-class decoding tasks). Significance: The resultssuggest that ConTraNet is robust to learn distinct features from different HMIparadigms and generalizes well as compared to the current state of the artalgorithms.

ConTraNet: A single end-to-end hybrid network for EEG-based and EMG-based human machine interfaces