Abstract

The overall goal of this dissertation is focused on addressing challenging problems in 1D, 2D/3D and 4D neuroimaging by developing novel algorithms that combine signal processing and machine learning techniques. One of these challenging tasks is the accurate localization of the eloquent language cortex in brain resection pre-surgery patients. This is especially important since inaccurate localization can lead to diminshed functionalities and thus, a poor quality of life for the patient. The first part of this dissertation addresses this problem in the case of drug-resistant epileptic patients. We propose a novel machine learning based algorithm to establish an alternate electrical stimulation-free approach, electro-corticography (ECoG) as a viable technique for localization of the eloqeunt language cortex. We process the 1D signals in frequency domain to train a classifier and identify language responsive electrodes from the surface of the brain. We then enhance the proposed approach by developing novel multi-modal deep learning algorithms. We test different aspects of the experimental paradigm and identify the best features and models for classification. Another difficult neuroimaging task is that of identifying biomarkers of a disease. This is even more challenging considering that skill acquisition leads to neurological changes. We propose to help understand these changes in the brain of chess masters via a multi-modal approach that combines 3D and 4D imaging modalities in a novel way. The proposed approaches may help narrow the regions to be tested in pre-surgical localization tasks and in better surgery planning. The proposed work may also pave the way for a holistic view of the human brain by combining several modalities into one. Finally, we deal with the problem of learning strong signal representations/features by proposing a novel capsule based variational autoencoder, B-Caps. The proposed B-Caps helps in learning a strong feature representation that can be used with multi-dimensional data.

Notes

If this is your thesis or dissertation, and want to learn how to access it or for more information about readership statistics, contact us at STARS@ucf.edu.

Graduation Date

2020

Semester

Summer

Advisor

Bagci, Ulas

Degree

Doctor of Philosophy (Ph.D.)

College

College of Engineering and Computer Science

Department

Computer Science

Degree Program

Computer Science

Format

application/pdf

Identifier

CFE0008589; DP0024265

URL

https://purls.library.ucf.edu/go/DP0024265

Language

English

Release Date

February 2022

Length of Campus-only Access

1 year

Access Status

Doctoral Dissertation (Campus-only Access)

Share

COinS