A hybrid approach based on analysis of hierarchies and decision tree methods to substitute treatment tactics for patients with atopic

The article considers the problem of choosing the optimal treatment strategy for patients with atopic dermatitis, which is multi-criteria and depends on clinical, individual and social factors. A hybrid approach is proposed that combines the Analytic Hierarchy Process (AHP) method to determine the significance weights of treatment criteria and a decision tree to model possible therapy scenarios and their results. The model allows systematizing the decision-making process, taking into account the uncertainty of the clinical response and the patient's personal priorities. In addition, for the task of differential diagnosis of atopic dermatitis forms, the XGBoost algorithm was used based on the analysis of 175 disease histories, which provided high classification accuracy (97.14%). The proposed approach contributes to increasing the validity and personalization of therapeutic decisions and can be used as the basis for clinical decision support systems in dermatology.