Detecting Breast Cancer from Thermal Infrared Images by Asymmetry Analysis
One of the popular methods for breast cancer detection is to make comparisons between contralateral images. When the images are relatively symmetrical, small asymmetries may indicate a suspicious region. In thermal infrared (IR) imaging, asymmetry analysis normally needs human intervention because of the difficulties in automatic segmentation. In order to provide a more objective diagnostic result, we describe an automatic approach to asymmetry analysis in thermograms. It includes automatic segmentation and supervised pattern classification. Experiments have been conducted based on images provided by Elliott Mastology Center (Inframetrics 600M camera) and Bioyear, Inc. (Microbolometer uncooled camera).