Surface temperature analysis and thermophysical property estimation for breast cancer by deep learning

Abstract

The breast cancer will affect the skin surface temperature profile, whose thermophysical properties are very important for the diagnose of the disease. The temperature on the skin surface is analyzed by a 3-D layered breast model with variable metabolic heat generations and blood perfusion rates. The thermophysical properties of the breast are estimated by deep learning. The relationship between the temperature profiles and the thermophysical properties of the cancer is revealed. The research shows that infrared thermography with deep learning is a useful diagnostic tool for the breast cancer.

Keywords:

• Breast cancer
• deep learning
• infrared thermography
• temperature profile
• thermophysical property estimation

Additional information

Funding

This work is supported by the National Natural Science Foundation of China, under Grant Nos. 12002181, 11921002, 11972205, 11672098 and 11722218.