Date of Award
Summer 2020
Document Type
Thesis
Degree Name
Master of Science (MS)
Department
Computer Science
Committee Chairperson
Richard Burns, Ph.D.
Committee Member
Si Chen, Ph.D.
Committee Member
Linh Ngo, Ph.D.
Abstract
This thesis presents a neural network system which segments images of the retina to calculate the cup-to-disc ratio, one of the diagnostic indicators of the presence or continuing development of glaucoma, a disease of the eye which causes blindness. The neural network is designed to run on commodity hardware and to be run with minimal skill required from the user by packaging the software required to run the network into a Singularity image. The RIGA dataset used to train the network provides images of the retina which have been annotated with the location of the optic cup and disc by six ophthalmologists, and six separate models have been trained, one for each ophthalmologist. Previous work with this dataset has combined the annotations into a consensus annotation, or taken all annotations together as a group to create a model, as opposed to creating individual models by annotator. The interannotator disagreements in the data are large and the method implemented in this thesis captures their differences rather than combining them together. The mean error of the pixel label predictions across the six models is 10.8%; the precision and recall for the predictions of the cup-to-disc ratio across the six models are 0.920 and 0.946, respectively.
Recommended Citation
Kilgannon, Jon, "A Machine Learning System for Glaucoma Detection using Inexpensive Machine Learning" (2020). West Chester University Master’s Theses. 172.
https://digitalcommons.wcupa.edu/all_theses/172