2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) 2019
DOI: 10.1109/embc.2019.8857575
|View full text |Cite
|
Sign up to set email alerts
|

Classification and Transfer Learning of EEG during a Kinesthetic Motor Imagery Task using Deep Convolutional Neural Networks

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1

Citation Types

0
4
0

Year Published

2020
2020
2024
2024

Publication Types

Select...
4
4
1

Relationship

1
8

Authors

Journals

citations
Cited by 10 publications
(4 citation statements)
references
References 18 publications
0
4
0
Order By: Relevance
“…The first design consideration is the number of convolutional layers, together with the type of end classifier. In MI tasks, 70% of CNN models use a rectified linear unit (ReLU) as the layer's activation function, while the vast majority of classifier fully connected layers employ a softmax activation function [49]. The proposed network relies on the Wide&Deep architecture for handling multiple inputs to learn deep patterns under simple rules.…”
Section: Discussion and Concluding Remarksmentioning
confidence: 99%
“…The first design consideration is the number of convolutional layers, together with the type of end classifier. In MI tasks, 70% of CNN models use a rectified linear unit (ReLU) as the layer's activation function, while the vast majority of classifier fully connected layers employ a softmax activation function [49]. The proposed network relies on the Wide&Deep architecture for handling multiple inputs to learn deep patterns under simple rules.…”
Section: Discussion and Concluding Remarksmentioning
confidence: 99%
“…The first design consideration is the number of convolutional layers, together with the type of end classifier. In MI tasks, 70% of CNN models use a rectified linear unit (ReLU) as the layers activation function, while the vast majority of classifier fully-connected layers employ a softmax activation function [46]. The proposed network is fully-connected that relies on the Wide&Deep architecture to handle multiple inputs to learn deep patterns under simple rules.…”
Section: A02tmentioning
confidence: 99%
“…While EEG measurements are affected by many factors, including physiological and non-physiological artifacts [3] resulting in low signal-to-noise ratios [4], recent advancements in de-noising (e.g., [5,6]) and deep learning [7,8] techniques have driven the emergence of viable clinical and non-clinical BCI applications based on scalp EEG [1,2,9]. These applications include, but are not limited to, seizure state prediction [10,11], sleep stage analysis [12], cognitive workload assessment [13], motor-imagery-based brain-computer interface (BCI) systems [14,15], neurorehabilitation [16,17], multi-modal and multi-brain-computer interfaces [18], braincontrolled vehicles [19], EEG-based home control [20], virtual reality [21], and interactive virtual environments [22]. While the future of these proof-of-concept BCI-enabled applications is promising, there are a number of technical challenges that remain before the widespread translation and adoption of these systems is realized.…”
Section: Introductionmentioning
confidence: 99%