Intelligent Support for Medical Decision Making

This paper presents the development and study of a model for formalizing the process of making a diagnosis using artificial intelligence methods. Currently, various artificial neural networks and expert systems have been created and are used for diagnosis. Analysis of these works has shown that these methods show good results, but have a number of drawbacks, the most significant of which is the complexity of organization and the significant time required to train a neural network. Thus, the problem is to develop new algorithms that have a probability of making an accurate diagnosis, comparable with artificial neural networks and expert systems, while having a shorter training time. One of the ways to solve this problem is to develop a model for diabetes diagnosis based on an artificial immune system. The purpose of this work is to develop and study of a model for formalizing the process of diagnosis using methods of artificial intelligence. The paper reviews a model of the diagnosis process: pre-diabetes (impaired glucose tolerance, impaired fasting glycemia), type I diabetes, type II diabetes. The problem of diagnosing the disease can be regarded as a classification problem. In this paper, the process of diagnosis was examined as a division of test data and patient history into four classes corresponding to one of the diagnoses: pre-diabetes (impaired glucose tolerance, impaired fasting glycemia), type I diabetes, type II diabetes. An artificial immune system and Kohonen artificial neural network were used to solve this problem.