We have developed flexible, active, multiplexed recording devices for high resolution recording over large, clinically relevant areas in the brain. While this technology has enabled a much higher-resolution view of the electrical activity of the brain, the analytical methods to process, categorize and respond to the huge volumes of data produced by these devices have not yet been developed. In our previous work, we applied advanced video analysis techniques to segment electrographic spikes, extracted features from the identified segments, and then used clustering methods (particularly Dirichlet Process Mixture models) to group similar spatiotemporal spike patterns. From this analysis, we were able to identify common spike motion patterns. In this paper, we explored the possibility of detecting and predicting seizures in this dataset using the Hidden Markov Model (HMM) to characterize the temporal dynamics of spike cluster labels. HMM and other supervised learning methods are united under the same framework to perform seizure detection and prediction. These methods have been applied to in-vivo feline seizure recordings and yielded promising results.