Date of Graduation

2016

Document Type

Thesis

Degree Type

MS

College

Statler College of Engineering and Mineral Resources

Department

Chemical and Biomedical Engineering

Committee Chair

Jeremy Dawson

Committee Co-Chair

Yuxin Liu

Committee Member

Linda Vona-Davis

Abstract

Hyperspectral Imaging (HSI) is a non-invasive optical imaging modality that shows the potential to aid pathologists in breast cancer diagnoses cases. In this study, breast cancer tissues from different patients were imaged by a hyperspectral system to detect spectral differences between normal and breast cancer tissues, as well as early and late stages of breast cancer. If the spectral differences in these tissue types can be measured, automated systems can be developed to help the pathologist identify suspect biopsy samples, which will improve sample throughput and assist in making critical treatment decisions. Tissue samples from ten different patients were provided by the WVU Pathology Department. The samples from each patient included both normal and ductal carcinoma tissue, both stained and unstained. These cells were imaged using a snapshot HSI system, and the spectral reflectances were evaluated to see if there was a measurable spectral difference between the various cell types. Analysis of the spectral reflectance values indicated that wavelengths near 550nm show the best differentiation between tissue types. This information was used to train image processing algorithms using supervised and unsupervised data. K-Means and Support Vector Machine (SVM) approaches were applied to the hyperspectral data cubes, and successfully detected spectral tissue differences with sensitivity of 85.45%, and specificity of 94.64% with TNR of 95.8%, and FPR of 4.2%. These results were verified by ground truth marking of the tissue samples by a pathologist. This interdisciplinary work will build a bridge between pathology and hyperspectral optical diagnostic imaging in order to reduce time and workload on the pathologist, which can lead to benefit of lead reducing time, and increasing the accuracy of diagnoses.

Download

Share

COinS