Seizure Types Classification by Generating Input Images With in-Depth Features From Decomposed EEG Signals for Deep Learning Pipeline.
Anand ShankarSamarendra DandapatShovan BarmaPublished in: IEEE J. Biomed. Health Informatics (2022)
Keyphrases
- eeg signals
- input image
- deep learning
- depth map
- signal processing
- extracted features
- brain computer interface
- unsupervised feature learning
- brain activity
- natural images
- eeg data
- unsupervised learning
- viewpoint
- face images
- gray level
- multi view
- machine learning
- superpixels
- weakly supervised
- segmentation algorithm
- saliency map
- spatial information
- feature set
- feature vectors
- training set
- pattern recognition
- image segmentation