Abstract

In this study, Doppler signals were recorded from 40 healthy volunteers and the right and left hand ulnar and radial arteries of 40 rheumatoid arthritis patients. Multiple Signal Classification method, one of the subspace signal processing methods is applied to the obtained Doppler signals and the feature of signs has been reached. Diseased and healthy people have been distinguished by using three different normalization techniques, including (z-score, minimum-maximum and decimal scaling) and artificial neural networks classification. K-fold cross-validation, classification accuracy, sensitivity and specificity are used to interpret and described the results of medical diagnostic test. Ali Osman Özkan"Effect of Normalization Techniques on Multilayer Perceptron Neural Network Classification Performance for Rheumatoid Arthritis Disease Diagnosis" Published in International Journal of Trend in Scientific Research and Development (ijtsrd), ISSN: 2456-6470, Volume-1 | Issue-6 , October 2017, URL: http://www.ijtsrd.com/papers/ijtsrd3576.pdf http://www.ijtsrd.com/engineering/electronics-and-communication-engineering/3576/effect-of-normalization-techniques-on-multilayer-perceptron-neural-network-classification-performance-for-rheumatoid-arthritis-disease-diagnosis/ali-osman-özkan

Authors and Affiliations