Mario Garingo
I use my knowledge in signal processing and machine learning techniques to solve various problems spanning from biological to audio and image applications
A Pilot Study of Epilepsy Seizure Structure Using Tensor Analysis and Source Localization with Sick Kids Hospital
◆ EEG ◆ Epilepsy ◆ Research ◆ Seizure ◆Project Lifespan: 2014
The purpose of this study was to identify the origins of epileptic seizures in children. My approach was to develop a model which incorporates source localized EEG current density vectors, frequency (through continuous wavelet transform) and the anatomical location. These three features generate a third order tensor with each node corresponding to a feature. In theory, this can be used for both seizure localization and seizure recognition, but for the purpose of this project I only investigated seizure localization. Though inconclusive due to the small number of patient samples, we demonstrated that tensor analysis and Parallel Factor Analysis can be used as a means of identifying complex structures of an epileptic seizure.