An integrated system for medical diagnosis: Clinical findings
Christos N. Schizas, Malek Fredj, Constantinos S. Pattichis, C. A. Bonnsee, K. Kyriallis, Lefkos T. Middleton · Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society · 1992
The diagnostic usefulness of artificial neural networks (ANNs) is explored by means of an integrated system for medical diagnosis. The models were developed by using the unsupervized self organizing feature maps algorithm. Clinical data for training and testing the ANNs was collected from 71 subjects by applying examination protocols that were developed by an expert neurologist. The diagnostic yield obtained by the examined models was in the region of 86 to 93%.