Creating a classification model for diagnosis of joint lesions type

摘要

The work combines methods of multidimensional polarization microscopy, statistical processing of data and algorithmsof machine learning with the purpose of constructing a methodology for creation of intelligent systems for multi-levelmedical monitoring of joint lesions . The task of classifying the results of the study of biological materials for obtaining adiagnosis was solved. To obtain informative features, a model of biological tissue was developed and the maindiagnostic parameters were determined (statistical moments of 1-4 orders of coordinate distributions of the values ofazimuths and the ellipticity of polarization and their autocorrelation functions, as well as wavelet coefficients of thecorresponding distributions). The classification of these data was provided on the raw input data and on generated datawith different degree of overlapping classes by machine learning algorithms and inductive modeling.