Temple University Hospital

Temple University Hospital's extensive EEG datasets, including the TUH EEG Corpus and its subsets, are driving significant advancements in automated seizure detection and classification. Research focuses on developing and optimizing machine learning models, such as convolutional neural networks (CNNs), recurrent neural networks (RNNs, including Bi-LSTMs), and transformer-based architectures (like BERT), to improve the accuracy and efficiency of EEG analysis. These efforts aim to reduce false positives, enhance sensitivity, and enable faster, more reliable diagnosis of epilepsy and different seizure types, ultimately improving patient care and reducing the burden on clinicians.