Spatio-Temporal-Spectral Hierarchical Graph Convolutional Network With Semisupervised Active Learning for Patient-Specific Seizure Prediction

MS4PS: A Mentor-Student Architecture for Patient-Specific Seizure Detection With Combination of Transfer Learning and Active Learning

“…There are many works that study patient-independent seizure detection, e.g., [1] employs a real-time method based on dictionary learning and sparse representation; [22] employs a method based on a deep neural network that combines a In order to adapt to the variation of seizure characteristics among patients and at different times, some patient-specific seizure detection algorithms have been proposed. Most of them suppose having enough labeled EEG data and try to improve the patient-specific detector's performance by using elaborate models and features, e.g., [2] uses a group of SVMs and features extracted through empirical mode decomposition (EMD) and common space patterns (CSP); [17] uses a voting SVM system and features containing both the temporal-domain and spectral-domain information of EEG; [23] uses a RVM model and the harmonic multiresolution and self-similarity-based fractal features from EEG data; [16] builds a predictor based on spatio-temporal-spectral hierarchical GCN with an active pre-ictal interval learning scheme (STS-HGCN-AL).…”

“…It makes the patientindependent seizure detection models hard to fit specific patients. Some research works [16]- [18] have been done and proved that building a patient-specific seizure detector for each patient seams more suitable than building a patientindependent seizure detector for all patients. Considering that it is still hard to train a suitable detector for a specific patient who has no historically-labeled data (e.g., when he/she sees a doctor for the first time) or has not enough historicallylabeled data, it is reasonable to employ transfer learning.…”

Active Learning Music Genre Classification Based on Support Vector Machine

“…In each iteration, after the samples within the geometric interval are selected from the unlabeled samples to join the training, the classification surface position of the new classifier is most likely to change, while the erefore, in the traditional SVM active learning method, it is considered that, for SVM classifier, those sample points closest to the optimal classification hyperplane are the most valuable sample points. However, if it is only used as the standard to judge whether the sample is the most valuable sample, it may face the problem of repeated learning, because the number of sample sets is often quite large in practice, so it is necessary to select the value samples in batch each iteration, and the value samples selected in batch each time are likely to have information redundancy due to the high correlation between them, which will lead to repeated learning [19]. In other words, we hope that the sample set selected by the classifier in each iteration not only has the most uncertainty, but also will maintain diversity, so as to maximize the modified classification hyperplane.…”

A Temporal-Spectral-Based Squeeze-and- Excitation Feature Fusion Network for Motor Imagery EEG Decoding

“…First, EEG electrodes are selected manually (i.e. HGD), which probably omits the spatial information of EEG sensors [35], and this neglecting of spatial dependencies in EEG signals may leads to the suboptimal decoding performance [36]. Therefore, our important future work is to apply adaptivelyselecting method which focuses on the most discriminative channels.…”