Neural Network and Artificial Immune Algori- thms for the Classification of Medical Data Series

This paper describes the applicability of artificial immunealgorithms. Medical data series classification technique by Artifi-cial Immune Algorithm is used for Neural Network Algorithminput data definitions. Artificial Immune Algorithms is created andtrained for the purpose of Arterial Blood Gas parameters classifica-tion: pH, PaCO2, PaO2, HCO3. The main goal of this paper is todevelop a artificial neural network technique for Arterial BloodGases short-term prediction. The main question that is considered ishow to predict some dynamic parameters that describe blood gasesnature. A model of a physical system has an error associated with itspredictions due to the dependences of the physical systems outputon uncontrollable and unobservable quantities. The use of artifi-cial methods creates the possibilities of obtaining some parametervalues on the proper level of probability. This would provide a di-rect feedback to the clinical staff about the progress of a patient, thesuccess of individual treatments, and quality of care as well as pre-dicting blood gas value