GENETICALLY FOUND, NEURALLY COMPUTED ARTIFICIAL FEATUHES WITH APPLICATIONS TO EPILEPTIC SEIZURE DETECTION AND PREDICTION

منو

" GENETICALLY FOUND, NEURALLY COMPUTED ARTIFICIAL FEATUHES WITH APPLICATIONS TO EPILEPTIC SEIZURE DETECTION AND PREDICTION "

Document Type	:	Latin Dissertation
Language of Document	:	English
Record Number	:	151222
Doc. No	:	ET23014
Main Entry	:	Hiram A. Firpi Cruz
Title Proper	:	GENETICALLY FOUND, NEURALLY COMPUTED ARTIFICIAL FEATUHES WITH APPLICATIONS TO EPILEPTIC SEIZURE DETECTION AND PREDICTION
Note	:	This document is digital این مدرک بصورت الکترونیکی می باشد
Abstract	:	This work presents the development of a new algorithm called Genetically Found,Neurally Computed (GFNC) features. The algorithm employs a genetic algorithm, aneural network. and a classifier in order to create the analytical expression of an artificialfeature, starting from raw features (original features given prior to the experiment), thathelp a classifier in its discrimination task. The algorithm was first tested on a problem ofdeciding whether two random vectors were parallel or not. Then, the algorithm wasapplied to the problem of epileptic seizure prediction and detection. Epileptic seizureprediction is a very complex problem since to date, seizures that characterize this disorderoccur suddenly without warning. Several-....,....-...,..tested for theQ1 PC1 bus cardBoth these projects mere sofixare des elopment efforts tonards contributing to dlfferentaspects of Roboucs and lZ1echatronics projects m the Controls and Roboucs Group..
Subject	:	Electericl tess
	:	برق
electronic file name	:	TL46243.pdf
Title and statement of responsibility and	:	GENETICALLY FOUND, NEURALLY COMPUTED ARTIFICIAL FEATUHES WITH APPLICATIONS TO EPILEPTIC SEIZURE DETECTION AND PREDICTION [Thesis]