Objective: To determine if properties of epileptic networks could be delineated using interictal spike propagation seen on stereo-electroencephalography (SEEG) and if these properties could predict surgical outcome in patients with drug-resistant epilepsy. Methods: We studied the SEEG of 45 consecutive drug-resistant epilepsy patients who underwent subsequent epilepsy surgery: 18 patients with good postsurgical outcome (Engel I) and 27 with poor outcome (Engel II-IV). Epileptic networks were derived from interictal spike propagation; these networks described the generation and propagation of interictal epileptic activity. We compared the regions in which spikes were frequent and the regions responsible for generating spikes to the area of resection and post-surgical outcome. We developed a measure termed source spike concordance, which integrates information about both spike rate and region of spike generation. Results: Inclusion in the resection of regions with high spike rate is associated with good post-surgical outcome (sensitivity = 0.82, specificity = 0.73). Inclusion in the resection of the regions responsible for generating interictal epileptic activity independently of rate is also associated with good post-surgical outcome (sensitivity = 0.88, specificity = 0.82). Finally, when integrating the spike rate and the generators, we find that the source spike concordance measure has strong predictability (sensitivity = 0.91, specificity = 0.94). Interpretations: Epileptic networks derived from interictal spikes can determine the generators of epileptic activity. Inclusion of the most active generators in the resection is strongly associated with good post-surgical outcome. These epileptic networks may aid clinicians in determining the area of resection during pre-surgical evaluation.
Anomaly anticipation in traffic scenarios is one of the primary challenges in action recognition. It is believed that greater accuracy can be obtained by the use of semantic details and motion information along with the input frames. Most state-of-the art models extract semantic details and pre-defined optical flow from RGB frames and combine them using deep neural networks. Many previous models failed to extract motion information from pre-processed optical flow. Our study shows that optical flow provides better detection of objects in video streaming, which is an essential feature in further accident prediction. Additional to this issue, we propose a model that utilizes the recurrent neural network which instantaneously propagates predictive coding errors across layers and time steps. By assessing over time the representations from the pre-trained action recognition model from a given video, the use of pre-processed optical flows as input is redundant. Based on the final predictive score, we show the effectiveness of our proposed model on three different types of anomaly classes as Speeding Vehicle, Vehicle Accident, and Close Merging Vehicle from the state-of-the-art KITTI, D2City and HTA datasets.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.