In this paper, a novel hybrid intelligent system (HIS) architecture is proposed. In contrast to the traditional sequence of initial analysis of incoming data in a knowledge base, the case under consideration first arrives for comparison in a library of analogs. If the new object is found to be highly similar to similar cases (based on a given metric), a hypothesis is accepted and a decision is issued to the user. In the absence of sufficient similarity, the requirement of which is set in the system, the solution of the problem is passed to the expert system for further processing using knowledge base rules. In this case, the accuracy of identification of the case unrecognized in the library of analogues can be increased due to greater consideration of vagueness, uncertainty and incompleteness in the initial data. The consequence of unsatisfactory recognition at this stage is the reference to the library of precedents (atypical cases). The proposed variant of the cascade procedure of object recognition provides higher efficiency of the system. This is especially important for diseases characterized by polymorphic clinical picture and with a progressive course of the pathological process. The realization of this approach is realized in the GIS prototype for diagnostics of Duchenne disease stages. The paper presents the results of system testing.
DOI: 10.1007/978-3-032-13612-1_29
Download the proceedings from Springer Nature: https://link.springer.com/content/pdf/10.1007/978-3-032-13612-1.pdf
ResearchGate: https://www.researchgate.net/publication/399128427_Hybrid_Intelligent_Medical_System_with_Decision-Making_Process_Modification
Kobrinskii, B. A., Nikolaev, A. A., Vlodavets, D. V. (2026). Hybrid Intelligent Medical System with Decision-Making Process Modification // Proceedings of the Ninth International Scientific Conference “Intelligent Information Technologies for Industry” (IITI’25), Volume 2, pp. 329–338.