Spectral Analysis of the Left and Right Hand Radial Artery Doppler Signals using the Welch Method to Diagnose Rheumatoid Arthritis Disease

In this study, 40 healthy control group and 40 patients with rheumatoid arthritis (RA) and Doppler signal from radial artery of the left and right hand were recorded. This non-parametric Welch Periodogram method to signal attributes of the signs were removed and the artificial neural network (ANN) classification method was utilized in order to recognize diseased and healthy control group individuals. Then, multilayer feed forward ANN trained with a Levenberg Marquardt (LM) back propagation algorithm was applied to lineament extracted from left and right hand Radial artery Doppler signals for classifying RA disease. The combination of Welch Periodogram method with ANN achieved the classification accuracies of 91.25 % left hand and 90 % right hand Radial artery Doppler signals in the diagnosis of RA disease. Radial artery left hand and right hand for the ROC curves and AUC values were calculated. DAS 28 for RA showing the extent of disease, classification of result values was correlated with each other. The suggested approach in this study is of potential to help with the early diagnosis of RA disease for the specialists who dealing with this subject. Keywords— Rheumatoid Arthritis, Radial Artery, Welch Periodogram Method, Artificial Neural Networks, Back Propagation Algorithm of LM