Performance Analysis Of Sparse Neural Networks In Brain Abnormality Detection

Author

• Megan Danh, University of Central FloridaFollow

Keywords

pruning; fmri; machine learning; sparse neural networks; ABIDE

Abstract

Neuroimages have held the capability of revealing to medical professionals patterns for brain abnormalities since their development. However, more recently, these professionals and researchers are looking to use neural networks to identify these brain abnormalities through neuroimages for early detection that would allow more effective treatment. Neuroimage datasets, specifically functional magnetic resonance imaging (fMRI), are extremely large in size. This would result in their processing and training to be computationally expensive, even with smaller neural networks. Fortunately, recent pruning methods have recently emerged, where network weights and neurons are pruned to reduce computational cost without compromising too much accuracy. By pruning multi-layer perceptron (MLP) that train on fMRI datasets, we can significantly reduce the power required to identify brain abnormalities in the image. In this thesis, we discuss how to apply various pruning methods towards this problem, and which method excels in performance.

Thesis Completion Year

2026

Thesis Completion Semester

Spring

Thesis Chair

Rahnavard, Nazanin

College

College of Engineering and Computer Science

Department

Department of Electrical and Computer Engineering

Thesis Discipline

Computer Engineering

Language

English

Access Status

Open Access

Length of Campus Access

None

Campus Location

Orlando (Main) Campus

STARS Citation

Danh, Megan, "Performance Analysis Of Sparse Neural Networks In Brain Abnormality Detection" (2026). Honors Undergraduate Theses. 565.
https://stars.library.ucf.edu/hut2024/565