Algorithms for the formation of two-dimensional characteristic and informative signs of diagnosis of diseases of the mammary glands by the methods of combined radio thermometry
Abstract:
Current paper is a part of group project to create a consultative information system that helps doctors to diagnose breast cancer with combined thermometry. For use in diagnostic algorithms, it is necessary to form a space of characteristic and highly informative diagnostic features. This paper outlines three algorithms for constructing two-dimensional features and presents the results of computational experiments that allow one to estimate the sensitivity and specificity of such features. Combining the best pairs of one-dimensional signs, it is possible to obtain the specificity and sensitivity of 70-80 %.
Keywords:microwave radiothermometry, data mining, breast screening, express diagnostics of malignant breast tumors, mammology.