An Algorithm for Detecting Artifacts in Video Recordings of Long-Term Video-EEG Monitoring Data for the Diagnostics of Delayed Cerebral Ischemia

In this paper, we propose an algorithm for detecting artifacts in long-term video-EEG monitoring data in the problem of diagnosing cerebral ischemia after subarachnoid hemorrhage. The algorithm is based on a threshold detector using the smoothed optical flow value. The optical flow is calculated from the video frames of long-term video-EEG monitoring. We conducted a computational experiment that showed the following: (a) artifacts are detected with accuracy acceptable for diagnosing cerebral ischemia during synchronous analysis of video data and EEG signals; (b) artifacts can be detected in real time.