PROGRAM PROCEDURES FOR TRAINING A RECOGNITION SYSTEM FOR THE DIFFERENTIAL DIAGNOSIS OF PATIENTS BASED ON HETEROGENEOUS SYMPTOM COMPLEXES

ISSN (p) 0321-2211, ISSN (e) 2663-3450 Автоматизація та інтелектуалізація приладобудування АВТОМАТИЗАЦІЯ ТА ІНТЕЛЕКТУАЛІЗАЦІЯ ПРИЛАДОБУДУВАННЯ DOI: 10.20535/1970.67(1).2024.306735 UDC 616.6:004.67 PROGRAM PROCEDURES FOR TRAINING A RECOGNITION SYSTEM FOR THE DIFFERENTIAL DIAGNOSIS OF PATIENTS BASED ON HETEROGENEOUS SYMPTOM COMPLEXES Shulyak O. P., Druzhynin V. V. National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute", Kyiv, Ukraine E-mail: , The machine learning recognition system for the differential diagnosis of patients based on heterogeneous nephrology parameter complexes is being considered, transitioning from instrumental means of examination. Training utilizes empirical statistics of clinical cases in a database with reliable diagnoses. The purpose is to expand the capabilities of information extraction from similar databases for training recognition procedures by enriching this toolkit with new features containing characteristic aspects of the extracted information. The research object is the mathematical and software toolkit for training recognition procedures of patient differential diagnosis based on statistics of reliably diagnosed clinical cases. The subject of the study is the software procedures for forming models of parameter complex incidence during training along scales of their values and the procedures for using these models in diagnostics. Model acquisition is perceived as the main content of the training process in ensuring diagnosis differentiation. A criterion for accepting preferential diagnostic decisions using such models is proposed. To simplify the development of mathematical and software procedures, heterogeneous symptom complexes are normalized and transformed to the [0; 1] scale. The introduction states the significant prevalence in medicine and related fields of databases with medical and biomedical data statistics on parameters and characteristics of human organs and systems in different conditions, their medical interpretation, and their use for various purposes, often associated with patient diagnostics. The problems of their formation and use are outlined on real databases, with one complicating factor in the development of diagnostic hardware-software being the substantial heterogeneity of parameters determined by patient examination instruments. Keywords: patient diagnosis; heterogeneous symptom complexes; parameter normalization; parameter distribution models; decision accumulation criterion. Introduction In both theory and practice of medicine and related fields, the prevalence of various open and closed-access databases of medical and biomedical data [1 – 3], diverse in their medical specialization and purpose, has become rooted and continues to progress. The extraction and utilization of information [1 – 3, 4] accumulated in such databases for various purposes, ranging from its study in professional training of specialists [1, 3, 5, 6] to its application in addressing various practical tasks in the field of medicine and related areas [1 – 4, 6 – 15], are gaining increasing relevance and importance. There remains a demand for the development of various software and hardware tools for obtaining necessary information from such databases in different sectors of subject area specialization [1, 3, 5, 16 – 18], including the demand for the development of simple specialized modules in software and hardware implementations [1, 3, 19 – 21]. Each type of toolkit for extracting necessary information from such databases and the corresponding tools that use it to address their issues have their own characteristics, their own emphases, and their effectiveness in extracting and using their available information [1 – 3, 9, 15, 21 – 26, 28] contained in the existing data, as well as their peculiarities in implementing components of accumulated empirical observation experience of objects, processes, and phenomena [1 – 4, 7, 22, 23, 26, 28] are of interest. Perhaps, there is no universal toolkit for such purposes, and each new development can be seen as obtaining data processing tools that complement the existing toolkit and may demonstrate sufficiently high effectiveness in their use, which needs to be verified for its effectiveness [1, 2, 5, 6, 21, 24, 26, 28], and in this sense, the relevance of such research and developments persists. One of the obstacles to the development of the mentioned simple specialized software and hardware data processing toolkit in the subject area under consideration is the heterogeneity of parameter complexes [1 – 6, 9, 10, 18, 22, 29, 30] collected in databases with descriptions of clinical cases. This complication can be overcome by simple uniform linear data transformations [41, 42] considered in the work. One of the main reasons for the heterogeneity of the mentioned databases is that they often represent collections of descriptions of clinical cases from medical practice with the results of patient instrumental Вісник КПІ. Серія ПРИЛАДОБУДУВАННЯ, Вип. 67(1), 2024. 55 ISSN (p) 0321-2211, ISSN (e) 2663-3450 Автоматизація та інтелектуалізація приладобудування examinations [1 – 3, 5 – 7, 12, 15] or the results of purposeful statistical studies [1 – 3, 10] related to the analysis of the impact and consequences of professional, climatic, and other conditions on human life processes [1, 2], the analysis of the dynamics of processes and phenomena in the body, the disclosure of relationships between the past, present, and future states of organs and systems at different levels in the body [1, 2, 4, 9, 10, 12 – 14, 19, 22], the identification of influencing factors [1, 7], risk factors [1, 22], chances of favorable outcomes [1, 18], as well as the determination of characteristic regional features [1, 9] related to population health provision, which explains the heterogeneity of parameter complexes in databases. Such databases contain real factual material of various physical nature, different levels of accuracy and reliability [1, 5, 6, 15, 28]. It is obtained empirically, including the use of software and hardware complexes of various, including medical, purposes and complexities, using unique and widely used means of patient examination, means of studying metabolic processes and products of human life activity, reactions to various influences, as well as tools for studying food products, water, determining environmental parameters, properties of biomedical materials [31 – 38]. Data may be collected during patient observation in the process of their dispensary examination, prevention and treatment, medical examinations, professional selection, surveys, categorization of the examined population by gender, age, working conditions, lifestyle, by risk groups and health level groups, by other characteristics as part of their comprehensive characterization [31 – 38]. This increases the diversity and heterogeneity of information in the obtained similar numerous parameter complexes in dat (...truncated)