Date of Award
8-2023
Document Type
Open Access Thesis
Degree Name
Master of Science (MS)
Department
Computer Science
First Advisor
Marc Pomplun
Second Advisor
Dan Simovici
Third Advisor
Ozgun Babur
Abstract
Epilepsy, a prevalent neurological disorder characterized by recurrent seizures, continues to pose significant challenges in diagnosis and treatment, particularly among children. Despite substantial advancements in medical technology and treatment modalities, localization of the part of brain that causes seizures (Epileptogenic Zone) remains a difficult task. Intracranial EEG (iEEG) is often used to estimate the epileptogenic zone (EZ) in children with drugresistant epilepsy (DRE) and target it during surgery. Conventionally, iEEG signals are inspected in the time domain by human experts aiming to locate epileptiform activity.
Visual scrutiny of the iEEG time-frequency (TF) images can be an alternative way to review iEEG allowing a detailed inspection in both the time and frequency domain. Though, this can be arduous for the human reader: subtle features of the TF image may be interictal indicators of the EZ that are not perceptible by the human eye.
This thesis focuses on an unsupervised, deep learning (DL) application of electrical brain signals (measured via intracranial EEG) from 32 paediatric patients (21 good outcomes and 11 poor outcomes with around 3,351 contacts) that we converted to Time-Frequency images (Morlet wavelet transform). We use VGG16 Deep Learning model to extract novel measures of image visual complexity in three different frequency bands that are of interest in the epileptic brain. Finally, we assessed the differences between signals that were generated in the surgical resection versus those outside (part of cortex surgically removed for seizure freedom) using Wilcoxon sign-rank test, separately in the patients who had a successful and unsuccessful surgery. The success of the surgery is based on whether the patient became seizure free after surgery (successful or good outcome) or continue to have seizures (unsuccessful or poor outcome). We aim to evaluate whether (and which of) the proposed Deep Leaning measures of visual complexity differed between resected and non-resected tissue but also depended on the surgery success.
Recommended Citation
Gupta, Sarvagya, "Visual Complexity of the Time-Frequency Image Pinpoints the Epileptogenic Zone: An Unsupervised Deep-Learning Tool to Analyze Interictal Intracranial EEG" (2023). Graduate Masters Theses. 783.
https://scholarworks.umb.edu/masters_theses/783
Included in
Bioinformatics Commons, Biomedical Engineering and Bioengineering Commons, Computer Sciences Commons