Degree Name



Electrical and Computer Engineering


Ira A. Fulton College of Engineering and Technology

Defense Date


Publication Date


First Faculty Advisor

Neal K. Bangerter

First Faculty Reader

David Wingate

Honors Coordinator

Karl Warnick


convolution, MRI, magnetic resonance, deep learning, neural network, cartilage


Recent advances in deep learning and convolutional neural networks (CNNs) have shown promise for automatic segmentation in magnetic resonance images. However, because of the stochastic nature of the training process, it is difficult to interpret what information networks learn to represent. This study explores multiple difference metrics between networks to determine semantic relationships between knee cartilage tissues. It explores how differences in learned weights and output activations between networks can be used to express these relationships. These findings are further supported by training multi-class networks to segment multiple tissues to compare network accuracy across different tissue combinations. This study shows that network generalizability for segmenting tissues can be measured by distances between networks. Femoral cartilage proves to be most closely related to other tissues, while patellar cartilage is most distant from other tissues. These findings are used to better understand feature extraction of knee cartilage in CNNs and to propose robust training policies for training semantic segmentation neural networks with limited data.