Intelligent decision support of diagnosis and therapy in nuclear medicine by Bayesian processing of uncertain data and probabilistic mixtures

First version of a decision support system using biophysical and radiological data was practically tested. Some of its insufficiencies indicate improvement of the methodology of mixture analysis and extension of the data set by biochemical, histological and other medical quantities. The aims:
1. Extension of theory and algorithms of Bayesian mixture analysis for consistent processing of discrete quantities and data with missing blocks. Result: the set of flexible software tools for implementation of mixture analysis as a method to get and manage information about a system using the system-generated data.
2. Collection and systemization of data from more patient examinations. Result:data set for the mixture analysis and for further use by doctors for research and clinic.
3. Development and implementation of the advisory module with algorithmic and software solutions of associated tasks. Result: software system for practical use.
4. Experimental run and testing of the advisory system.