THE ROLE OF ENDOSCOPIC ULTRASONOGRAPHY IN DIFFERENTIATING OBSTRUCTIVE PATHOLOGIES OF THE DUODENUM
##plugins.themes.academic_pro.article.main##
Abstract
Objective: to improve the quality of endoscopic ultrasonography (EUS) in the diagnosis of obstructive pathology of the major duodenal papilla (MDP) using fuzzy diagnostic models. Materials and methods. The study included 189 patients treated at the Kursk Regional Multidisciplinary Clinical Hospital in 2013-2023 . 134 people (70.9%) were diagnosed with stenosis of the BDJ, 24 (12.7%) were diagnosed with single benign tumors of the papilla, and 31 (16.4%) patients were diagnosed with malignant tumors of the BDJ. Based on the analysis of treatment results, the possibility of differential diagnosis of BDK pathology based on EUS data using fuzzy diagnostic models was studied. Differentiation was carried out using the methodology of synthesis of hybrid fuzzy decision rules, within which the lack of statistical data is filled with the clinical knowledge of experts, which determines the possibility of constructing a membership function for the pathology being differentiated. Results. As part of the methodology, informative echographic criteria for diagnosing various pathologies of the BDK based on endosonography data have been developed , systematized and presented in the form of standard endosonograms and reference diagrams, which reflect the characteristics of the diagnostic object as fully as possible. Based on expert assessment data, confidence coefficients in decisions made were determined, which are the basis of the diagnostic algorithm. The results of mathematical modeling showed confidence in the correct diagnosis of pathological changes using synthesized fuzzy algorithms over 0.85, which is an acceptable result for clinical use. Conclusion. The use of the developed fuzzy model in combination with visual inspection of the papillary zone (which is provided by the EUS method) brings the diagnostic accuracy of differential diagnosis closer to maximum values.
##plugins.themes.academic_pro.article.details##
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.