This work was carried out as part of research on the development of methods for the intelligent analysis of medical thermometric data. These methods are designed to create a consultative intelligent system for the diagnosis of breast cancer. In this work, an artificial neural network has been obtained, which allows localization of a tumor according to microwave thermometry data. The network uses the architecture of deep belief networks. In cases where the tumor does not coincide and is not near the hottest point of the breast, the neural network can reach an accuracy of over 56% on test samples. This result exceeds the previous one by 13%.
Keywords: microwave thermometry, breast cancer, mammology, tumor localization, fidelity, data mining, artificial neural networks, deep learning, thermometric diagnostic features, cross-validation