The objective of this exploratory study was to develop signal processing methods for assisting in the diagnosis of arteriovenous fistula stenosis on patients suffering from endstage renal disease and undergoing haemodialysis treatments. The proposed method is based on the classification of vessels sounds utilizing parameter extraction from wavelets transform coefficients. The coefficients energy of selected scales (frequency bands) were fed to a support vector machine based system for classification. Results suggested that this technique can be useful for diagnosis purposes to physicians during the auscultation procedure.

The Empirical Mode Decomposition (EMD) is a method to decompose non linear, non stationary time series into a sum of different modes, named Intrinsical Mode Functions each one having a characteristic frequency. In the present work we used the EMD to investigate the properties of the recorded sounds from the Arteriovenous fistula on hemodialysis patients. Phonoangiographic signals coming from two different vessel conditions, stenotic and non-stenotic, were analyzed by using EMD, the mean energy and mean instantaneous frequency per IMF proved to be good features for classification. Three types of classification schemes were tested on data from the first IMf features achieving good results.