Understanding how the human brain works is essential for gaining insights into behavior, for identifying brain disorders, and for improving patient care. Intracranial electroencephalography (iEEG) is a tool for studying brain activity. Recordings from iEEG sometimes show transient events that may be correlated with brain functions and certain neurological disorders. Reliable detection of these events is important for clinical applications and advancing neuroscience. This project will develop an artificial intelligence (AI)-based tool to identify events in iEEG recordings. The tool will detect known events and enable the identification of previously unrecognized brain events. The outcomes of this project could improve treatments for patients suffering from such disorders as epilepsy and Parkinson’s disease. In addition, the project will provide training opportunities for students and clinicians. Overall, the project will result in new AI-based healthcare technologies and contribute to a skilled workforce in applied AI. This Engineering Research Initiation project will provide improved tools for automated iEEG analysis. Most existing event detectors identify only a single type of event, without knowledge of other events, which can lead to misclassifications. The majority of available iEEG data are obtained from patients and tagged by expert annotations. This process is time-intensive and prone to expert subjectivity. To address these challenges, this project will develop a deep learning-based model trained on expert-annotated iEEG data from large, multicenter human datasets. The model will be evaluated by applying a leave-one-institution-out approach. Data from a single center will be left out as the test set. The training will be performed on data from the remaining centers. Then, the model will be evaluated on the data from the left-out center. Finally, the detector will be implemented as user-friendly, open-source software. This approach will ensure disseminat