Multidimentional thermometric data mining in medical diagnostics

This paper is devoted to the development of technology advisory intelligent systems and artificial intelligence methods intended for analysis, modeling and interpretation of medical thermometer data. During the last decade microwave radiometry method is widely used in various fields of medicine. However, the currently existing diagnostic system is a system of decision support for a very qualified person. This feature greatly narrows the scope and, in many respects, eliminates the possibilities of the method in the early diagnosis. The solution to this problem provides the development of an intellectual advisory system, that is expert system comprising a mechanism explanation and justification of the proposed solutions in a language understandable to the user. In this paper we propose a new approach to intellectual analysis of microwave radiometry data intended for the diagnosis of breast cancer. We present the method of formation of multidimensional information signs on the basis of a quantitative description of the existing medical knowledge and acquire new knowledge on the basis of physical and mathematical models of temperature fields. The resulting set of diagnostic features is the basis for the establishment of consultative Intellectual cancer diagnosis system of breast. In addition, the proposed method provides a similar diagnostic systems of a number of other diseases based on microwave radiometry data.