Unified Convolutional Sparse Transformer for Disease Diagnosis, Monitoring, Drug Development, and Therapeutic Effect Prediction from EEG Raw Data

• Published in: Biology Vol. 13; no. 4; p. 203
• Main Authors: He, Zhengda, Chen, Linjie, Xu, Jiaying, Lv, Hao, Zhou, Rui-ning, Hu, Jianhua, Chen, Yadong, Gao, Yang
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 22.03.2024
• Subjects: Accuracy; Artificial intelligence; Biological Sciences; Biology (General); Brain research; Computational linguistics; Computational neuroscience; Convulsions & seizures; Data analysis; data collection; Deep learning; Design; Diagnosis; Disease; disease diagnosis; Drug development; Drug discovery; Drugs; EEG; EEG analysis; Electric transformers; Electroencephalography; Epilepsy; Health aspects; Language processing; Machine learning; Medical diagnosis; Mental depression; Natural language interfaces; Neglect syndromes; Neural networks; Neurophysiology; Neurosciences; Precision medicine; prediction; psychiatric drug discovery; Psychotropic drugs; QH301-705.5; sparse transformer; Statistical analysis; statistics; Temporal variations; therapeutics; Wavelet transforms; Taiwan; disease diagnosis; psychiatric drug discovery; sparse transformer; EEG analysis
• ISSN: 2079-7737; 2079-7737
• DOI: 10.3390/biology13040203

Electroencephalogram (EEG) analysis plays an indispensable role across contemporary medical applications, which encompasses diagnosis, monitoring, drug discovery, and therapeutic assessment. This work puts forth an end-to-end deep learning framework that is uniquely tailored for versatile EEG analysis tasks by directly operating on raw waveform inputs. It aims to address the challenges of manual feature engineering and the neglect of spatial interrelationships in existing methodologies. Specifically, a spatial channel attention module is introduced to emphasize the critical inter-channel dependencies in EEG signals through channel statistics aggregation and multi-layer perceptron operations. Furthermore, a sparse transformer encoder is used to leverage selective sparse attention in order to efficiently process long EEG sequences while reducing computational complexity. Distilling convolutional layers further concatenates the temporal features and retains only the salient patterns. As it was rigorously evaluated on key EEG datasets, our model consistently accomplished a superior performance over the current approaches in detection and classification assignments. By accounting for both spatial and temporal relationships in an end-to-end paradigm, this work facilitates a versatile, automated EEG understanding across diseases, subjects, and objectives through a singular yet customizable architecture. Extensive empirical validation and further architectural refinement may promote broader clinical adoption prospects.