MULTI-AGENT TECHNOLOGY IN THE EARLY DIAGNOSIS OF SKIN MELANOMA

Abstract

The problems of the use of existing forms of organization in health care, focused on timely diagnosis of tumors of external localization, are analyzed. The urgency of the problem and the need to improve its effectiveness are determined by a number of factors, including numerous publications in Russia and abroad, as well as many years of experience of the authors. An approach based on interdisciplinary technology (medicine, software engineering) using multi-agent methods is proposed. The concept of technology and the model of the system composition consisting of an artificial neural network, an expert system, a database, a knowledge base, remote access means and information protection were developed. As the criteria for the effectiveness of the system, entropic actions of various groups of users are offered - doctors of profile and non-profile specialties in the space of situations. The end result of the diagnosis is survival. The proposed technology is applied in the educational process, designed for a wide range of users, including students, doctors, administrative staff, and average medical personnel. A methodology has been developed that contains theoretical and practical material intended for an extended version of the analysis of errors and recognition of complex scenes possible in the diagnosis of skin melanoma.