Machine learning methods for diagnosis of epilepsy from EEG

• 主要作者: Roshini, Koppala
• 其他作者: Justin Dauwels
• 格式: Theses and Dissertations
• 語言: English
• 出版: 2018
• 主題: DRNTU::Engineering::Electrical and electronic engineering
• 在線閱讀: http://hdl.handle.net/10356/76334
• 機構: Nanyang Technological University

Epilepsy is a neurological disorder presented with unpredicted and repeated seizures due to abnormal electrical activity in the brain. They can be diagnosed by analysing the electroencephalogram (EEG), which shows spikes when there is epileptic activity. The aim of this dissertation; “MACHINE LEARNING METHODS FOR DIAGNOSIS OF EPILEPSY FROM EEG”, is to develop a generic system which will be able to predict if the patient is epileptic. The system is built using Machine Learning algorithms, like k-Nearest Neighbour, Neural Networks and Convolutional Neural Networks. The algorithm is trained on interictal scalp EEG data recorded from epileptic patients. The project is in collaboration with neurologists at Massachusetts General Hospital and Harvard Medical School and applied Mathematicians at MIT. Once the EEG of the patient is fed to the system it should go through stream of independent processes and finally assess if the patient is potentially positive for epilepsy.