DEVELOPMENT OF A METHODOLOGY FOR UTILIZING INFRARED IMAGING TO ANALYZE THERMAL RESPONSE AND DIAGNOSE BREAST CANCER

Abstract

Breast cancer is among the diseases that kill the most women in Brazil and worldwide. The lethality of the disease is related to its stage, that is, to the degree of the disease's involvement in the individual. Early diagnosis is, therefore, of utmost importance to increase the chances of cure and patients' survival. The thermography exam, a safe method free of radiation exposure and physical contact, is capable of detecting the first metabolic alterations caused by a tumor. However, it provides only metabolic information and not anatomical information about the tumor, besides being strongly influenced by environmental factors (humidity, radiation, room temperature) and patient factors (age, breast shape and size). The objective of the work is, therefore, to develop a standardized method of using thermography through infrared imaging for the diagnosis of breast cancer. The method aims to provide anatomical information about the tumor, based on metabolic information from infrared imaging. For this purpose, breast imaging exams will be transformed into 3D STL models and subsequently smoothed and discretized using a uniform cubic mesh. The Method of Volume Elements divides the domain to be studied into control volumes, with each element of the mesh corresponding to a volume element of the breast. Using physical laws and empirical and theoretical correlations for mass, heat, and fluid flow, each volume element can be represented by a system of ordinary differential equations, which indicate the heat exchange and blood flow in each unit. In this way, the internal and surface temperature distributions of the breast as functions of time, space, and known initial and boundary conditions can be calculated. By comparing the obtained surface temperature of the breast and the actual infrared image temperature distribution of the breast, it will be possible to simulate the internal temperature distribution of the breast and obtain a precise estimate of the tumor location. This approach could make thermography more independent of complementary exams, as well as enable accurate early-stage cancer diagnosis and tumor depth prediction.