PREDICTION OF HEMORRHAGIC BLOOD-LOSS WITH A GENETIC ALGORITHM NEURAL-NETWORK

There is no established method for accurately predicting how much bloo d loss has occurred during hemorrhage. In the present study, we examin e whether a genetic algorithm neural network (GANN) can predict volume of hemorrhage in an experimental model in fats and we compare its acc uracy to stepwise linear regression (SLR). Serial measurements of hear t period; diastolic, systolic, and mean blood pressures; hemoglobin; p H; arterial PO2; arterial PCO2; bicarbonate; base deficit; and blood l oss as percent of total estimated blood volume were made in 33 male Wi star rats during a stepwise hemorrhage. The GANN and SLR used a random ly assigned training set to predict actual volume of hemorrhage in a t est set. Diastolic blood pressure, arterial PO2,, and base deficit wer e selected by the GANN as the optimal predictors set. Root mean square error In prediction of estimated blood volume by GANN was significant ly lower than by SLR (2.63%, SD 1.44, and 4.22%, SD 3.48, respectively ; P < 0.001). A GANN can predict highly accurately and significantly b etter than SLR volume of hemorrhage without knowledge of prehemorrhage status, rate of blood loss, or trend in physiological variables.