An empirical correlation to estimate thermal properties of the breast and of the breast nodule using thermographic images and optimization techniques

The thermography has been used as a complementary diagnostic technique in the breast cancer detection, mainly in young patients. For these patients, the mamography is not suitable due to the dense characteristic of their breasts. Infrared imaging has also been used to validate 3D numerical simulations, which aim to reach a better understanding of breast abnormalities. However, to build a more realistic model, it is necessary to know the thermophysical properties of the various types of breast tissues and of the various possible breast disorders (malignant, benign and cyst). The greatest difficulty to calculate the temperature profiles is caused by the uncertainty about the correct values of the thermophysical properties mentioned above. This paper presents a methodology for estimating thermal properties based on the breast infrared images. An empirical correlation relating the breast thermal conductivity with the patient's age is also proposed. The maximum temperature measured by the thermography on the breast surface region over the abnormalities was used to estimate the thermophysical properties of the breast and of the tumor by solving an inverse problem. The CFD (Computational Fluid Dynamics) commercial software FLUENT, which uses the Finite Volume Method, was adopted for the numerical simulation. The Sequential Quadratic Programming (SQP) was used to solve the inverse heat transfer problem to estimate the thermal conductivity and the blood perfusion of breast tissues and tumors. The results obtained by the use of the proposed methodology showed that it is possible to estimate the thermophysical properties using infrared images. Furthermore, the empirical correlation relating the breast thermal conductivity with the patient's age provided satisfactory response for the analyzed cases. (C) 2019 Elsevier Ltd. All rights reserved.