This work includes a brief overview of the applications of the powerful and easy-to-perform method of microwave radiometry (MWR) for the diagnosis of various diseases. The main goal of this paper is to develop a method for diagnosing breast oncology based on machine learning algorithms using thermometric data, both real medical measurements and simulation results of MWR examinations. The dataset includes distributions of deep and skin temperatures calculated in numerical models of the dynamics of thermal and radiation fields inside biological tissue. The constructed combined dataset allows us to explore the limits of applicability of the MWR method for detecting weak tumors. We use convolutional neural networks and classic machine learning algorithms (k-nearest neighbors, naive Bayes classifier, support vector machine) to classify data. The construction of Kohonen self-organizing maps to explore the structure of our combined dataset demonstrated differences between the temperatures of patients with positive and negative diagnoses. Our analysis shows that the MWR can detect tumors with a radius of up to 0.5 cm if they are at the stage of rapid growth, when the tumor volume doubling occurs in approximately 100 days or less. The use of convolutional neural networks for MWR provides both high sensitivity ($sens=0.86$) and specificity ($spec=0.82$), which is an advantage over other methods for diagnosing breast cancer. A new modified scheme for medical measurements of IR temperature and brightness temperature is proposed for a larger number of points in the breast compared to the classical scheme. This approach can increase the effectiveness and sensitivity of diagnostics by several percent.

This article presents the development of an anatomical breast phantom for investigating the capabilities of microwave radiometry in assessing thermal processes in biological tissues. The phantom accounts for the heterogeneous tissue structure and haemodynamics, enabling realistic heat transfer modelling. Numerical simulation software was developed, accurately reproducing experimental results and allowing the study of thermal anomalies. Experimental validation demonstrated that the temperature in the subcutaneous layer differed on average by 0.3 °C from deeper tissues, confirming the method’s effectiveness. The presence of a tumour in the model resulted in a local temperature increase of up to 0.77 °C, highlighting the sensitivity of microwave radiometry to tumour-induced thermal anomalies. These findings contribute to enhancing non-invasive techniques for early breast disease detection.

This work was done with the aim of developing the fundamental breast cancer early differential diagnosis foundations based on modeling the spacetime temperature distribution using the microwave radiothermometry method and obtained data intelligent analysis. The article deals with the machine learning application in the microwave radiothermometry data analysis. The problems associated with the construction mammary glands temperature fields computer models for patients with various diagnostics classes, are also discussed. With the help of a computer experiment, based on the machine learning algorithms set (logistic regression, naive Bayesian classifier, support vector machine, decision tree, gradient boosting, Knearest neighbors, etc.) usage, the mammary glands temperature fields computer models set adequacy.

This work was carried outaspartof research on the development of methods forthe intelligent analysis of medical thermometric data. These methods are designed to create a consultative intelligent system for the diagnosis of breast cancer. The previously developed approach to the formation of the feature space based on microwave radiothermometry data and the created algorithm forthe localization of malignant neoplasms in the mammary gland were applied. This algorithm is a weighted voting algorithm that is configured using a genetic algorithm. The operation of the algorithm was tested separately on the results of modeling the temperature fields of the mammary glands, as w’cll as on real data. In addition, it was crosschecked against real data and simulation results. In these cases, the localization algorithm can achieve accuracy from 55 to 65% on test samples.