Abstract

Infantile Spasms (ISS) characterized by electroencephalogram (EEG) recordings exhibiting hypsarrythmia (HYPS) are a severe form of epilepsy. Many clinicians have been trying to improve ISS outcomes; however, quantification of discharges from hypsarrythmic EEG readings remains challenging.

This thesis describes the development of a novel method that assists clinicians to successfully localize the epileptic discharges associated with ISS in HYPS. The approach includes: construct the time-frequency domain (TFD) of the EEG recording using matching pursuit TFD (MP-TFD), decompose the TFD matrix into two submatrices using nonnegative matrix factorizations (NMF), and employ the decomposed vectors to locate the spikes.

The proposed method was employed to an EEG dataset of five ISS individuals, and identification of spikes was compared with those which were identified by the epileptologists and those obtained using clinical software (Persyst). Performance evaluations showed results based on classification techniques: thresholdings, and support vector machine (SVM). Using the thresholdings, average true positive (TP) and false negative (FN) percentages of 86% and 14% were achieved, which represented a significant improvement over the use of Persyst, which only achieved average TP and FN percentages of 4% and 96%, respectively. Using SVM, the percentage of area under curve (AUC) of receiver operating characteristic (ROC) was significantly improved up to 98.56%.

In summary, the proposed novel algorithm based on MP-TFD and NMF was able to successfully detect the epileptic discharges from the dataset. The development of the proposed automated method can potentially assist clinicians to successfully localize the epileptic discharges associated with ISS in HYPS. The quantitative assessment of spike detection, as well as other features of HYPS, is expected to allow a more accurate assessment of the relevance of EEG to clinical outcomes, which is significant in therapy management of ISS.

Library of Congress Subject Headings

Epilepsy in children--Data processing; Electroencephalography--Data processing

Publication Date

9-2015

Document Type

Thesis

Student Type

Graduate

Degree Name

Electrical Engineering (MS)

Department, Program, or Center

Electrical Engineering (KGCOE)

Advisor

Behnaz Ghoraani

Advisor/Committee Member

Daniel Phillips

Advisor/Committee Member

Sohail A. Dianat

Comments

Physical copy available from RIT's Wallace Library at RJ496.E6 T73 2015

Campus

RIT – Main Campus

Plan Codes

EEEE-MS

Download

Share

COinS