Files in this item



application/pdfECE499-Fa2008-parawira.pdf (2MB)Restricted to U of Illinois
(no description provided)PDF


Title:Breast Cancer Detection through Hyperspectral Image Analysis and Machine Learning
Author(s):Parawira, Sander; Brady, Spencer
Contributor(s):Do, Minh N.
Subject(s):cancer detection
breast cancer detection
medical imaging
image analysis
hyperspectral image analysis
Fourier transform infrared spectroscopy
Abstract:In this paper, we present a machine learning technique in hyperspectral image analysis for breast cancer detection. We consider the scenario where a band sequential image of a breast tissue specimen is collected using Fourier transform infrared (FTIR) spectroscopy. Our goal is to determine which parts of the image correspond to epithelium cells and which parts of the image correspond to stoma cells. We first discuss spectrum (band) selection and Principal Component Analysis (PCA) to reduce the dimensionality of the image. We then discuss Euclidean Distance and implemention of K-Means Clustering algorithm for data set represented as points in a high dimensional space. We finally add noise to the image and discuss the robustness of our technique.
Issue Date:2008-12
Genre:Dissertation / Thesis
Publication Status:unpublished
Peer Reviewed:not peer reviewed
Date Available in IDEALS:2014-03-12

This item appears in the following Collection(s)

Item Statistics