Method study hemostasis parameters using the neural network technology, comprising a stepwise (hierarchical) analysis of blood chemistry parameters with separation of normal parameters, and further differentiating pathologies, characterized in that the analysis parameters is performed by software; investigated hemostatic parameters are assigned weights (based on neural network learning); it is possible to configure a specific pathology hemostasis system, or duplication of one network level hierarchy to solve the problem of differentiation of a wide range of pathologies hemostasis system pathology separation stages and disease outcomes (with fine adjustment of each of the second level neural network for a specific pathology, step or outcome); it is possible to simultaneously analyze a group of patients in all investigated indices; It does not require the calculation of statistical indicators to assess the results of the analysis supplies to a particular class; neural network training is on real clinical data.
展开▼
