UTILIZATION OF A NEURAL-NETWORK IN THE ELABORATION OF AN EVALUATION SCALE FOR PAIN IN CEREBRAL-PALSY

An interesting aspect of neural networks is shown in the elaboration o f an evaluation scale for pain in cerebral palsy with severe mental re tardation. Because of the diversity of cases, the number of items had to be limited in the final step of statistical validation. Classical a nalysis on prior data did not allow to decide whether the variability in results is more likely due to the type of disability (i. e., the po ssibility of pain expression) than to the actual presence of pain. A n eural network was used to find implicit relations between the data, wi th the advantage of having total control on the variables' status by a pplying variations in the network architecture. This allowed for the r apid identification more significant item combinations as a function o f degree of relationship to pain in cerebral palsy.