Next Generation Hybrid Intelligent Medical Diagnosis Systems

Abstract

Many medical diagnosis problems (MDPs) are difficult to be solved by physicians. Frequently, the difficult MDPs solving by physicians require the assistance of the medical computing systems, which many times should be intelligent. Many papers presented in the specialized literature prove, that the intelligence of a system (frequently agent-based) can offer advantages in the MDPs solving versus a system that does not have such intelligence. Cooperative hybrid (human-machine) medical diagnosis systems seem to be well suited for the solving of many difficult MDPs. A difficult aspect in the design of such systems consists in the establishment of how to combine in an optimal way the humans and intelligent systems interoperation in order to solve the undertaken problems in the most efficient way. With this purpose, a novel hybrid medical system, called Intelligent Medical Hybrid System (IntHybMediSys) is proposed in this paper, a system which combines efficiently the humans and computing systems advantages in the problem-solving. We give a definition to the Difficult Medical Diagnosis Problem Solving Intelligence. IntHybMediSys is a highly complex hybrid system composed of physicians and intelligent agents that can interoperate intelligently in different points of decision in order to solve efficiently very difficult medical diagnosis problems. IntHybMediSys is able to handle emergent information that rise during the medical problems solving that allows the precise establishment of the most efficient contributor (a physician or an artificial agent) at each contribution during a problem-solving. This kind of problem-solving has as an effect the increase of accuracy of the elaborated diagnostic.