Assessment of Neurological Function with Multimodal and Multichannel Physiological Signal Analysis Using Machine and Deep Learning Techniques

This study explored how combining EEG (a measure of the electrical activity of the brain) and ECG (a measure of electrical activity of the heart) with artificial intelligence (AI) could enhance the monitoring of brain function in patients with delirium or those under general anesthesia. By integrating data from these two modalities and applying advanced AI techniques, we aimed to achieve a clearer understanding of a patient’s neurological condition. We focused on several key steps of signal analysis: cleaning the data from artifacts, merging information from EEG and ECG, and classifying the results. Our distinctive approach involved mapping both brain and heart signals into a unified space, which allowed us to identify complex patterns associated with above mentioned medical conditions. Although our method successfully classified neurological states, it did not reveal which specific abnormalities in EEG or ECG influenced the classification. Further research is needed to address this limitation.