Abstract:
Breast cancer is the second leading cause of death in women worldwide with an average probability for a woman to develop breast cancer in her life of about 12%. Among the large variety of medical assessment techniques, thermography has attracted attention in applications related to detection and diagnosis due to its capability to provide valuable information on the physiological variations typical of early stages in cancer development thus making possible to diagnose patients in early stages so more thorough examinations can be done in proper time and manner. This paper presents a digital processing approach that allows identification and subsequent isolation of the region of interest in thermograms based texture analysis of the image. This algorithm was tested on case studies thermograms exhibiting different types of cancer and the results showed successful identification and extraction of the region of interest in all cases. Results are presented with different types of cancer in men and women and different image angles showing the robustness of the proposed method
