Date Approved

2-13-2024

Embargo Period

2-13-2024

Document Type

Thesis

Degree Name

Master of Science (M.S.)

Department

Biomedical Engineering

College

Henry M. Rowan College of Engineering

Advisor

Erik Brewer, Ph.D.

Committee Member 1

Mary Staehle, Ph.D.

Committee Member 2

Nidhal Bouaynaya, Ph.D.

Committee Member 3

Shen-Shyang Ho, Ph.D.

Keywords

Bone Marrow Lesion, Image Regression, Machine Learning

Subject(s)

Osteoarthritis; Image processing; Neural network (Computer science)

Disciplines

Biomedical Engineering and Bioengineering | Computer Sciences | Medicine and Health Sciences

Abstract

Bone marrow lesions (BMLs), occurs from fluid build up in the soft tissues inside your bone. This can be seen on magnetic resonance imaging (MRI) scans and is characterized by excess water signals in the bone marrow space. This disease is commonly caused by osteoarthritis (OA), a degenerative join disease where tissues within the joint breakdown over time [1]. These BMLs are an emerging target for OA, as they are commonly related to pain and worsening of the diseased area until surgical intervention is required [2]–[4]. In order to assess the BMLs, MRIs were utilized as input into a regression convolutional neural network (CNN). Initial experimentation handled the MRI using each individual slice in a 2D convolutional neural network as a baseline model, eventually progressing to a tensor stacked input into a 3D convolutional neural network. The viability and effectiveness of the models were evaluated using mean absolute error (MAE), mean squared error (MSE), root mean squared error (RMSE), and R2. The 3D regression CNN models were observed to perform better than the 2D regression CNN models.

Download

Share

COinS