SIMULATION OF OXYGEN-SATURATION OF HEMOGLOBIN SOLUTION, RBC SUSPENSION AND HEMOSOME BY A NEURAL-NETWORK SYSTEM

Hemoglobin-based artificial blood substitutes as oxygen carrier is adv antageous over current plasma expander. In this study, oxygen saturati on of hemoglobin solution, red blood cell suspension and artificial bl ood substitute under various conditions were measured by yeast-consumi ng-oxygen experiments instead of spectrophotometer. The empirical resu lts were assigned into training feedforward back-propagation neural ne twork system in order to simulate the oxygen saturation model modulate d by those factors such as pH, [Cl-], [2,3-DPG], pO(2) and pCO(2). Con sequently, this neural network is able to simulate accurately the oxyg en saturation of Hb solution. The prediction of hemosome is not agreed well possibly because of the resistance of transport of oxygen. Howev er, the results showed neural net can offer a simple and convenient wa y in comparison with the conventional methods, especially in dealing w ith complex and ambigious problem.