Untitled

Abstract

This paper proposes a signal processing method for the evaluation of arteriovenous (AV) shunt stenosis in hemodialysis patients. AV shunts are surgically created pathological fistulas that serve as access routes for hemodialysis. The distinct and periodic bruit of the vascular shunts is clearly audible over the access routes. Thus, a bruit spectral analysis can be a valuable noninvasive method for quantifying the severity of vascular stenosis. This study collected phonoangiographic data from thirty AV shunts obtained from an electronic stethoscope during pre-and post-percutaneous transluminal angioplasty (PTA) periods. An autoregressive (AR) model was applied to analyze the phonoangiographic signals. The AR model and a filter order of eight were chosen to estimate the characteristic frequency of the bruit. AR model results obtained from the analysis of the phonoangiographic data under the pre-and post-PTA conditions show significant changes in frequency and magnitude. Seven patients were enrolled for periodical follow-up analysis for AV shunts. The Burg AR model is used to find the characteristic frequency of phonoangiographic signals. Therefore, the variation of frequency and amplitude in power spectra analysis showed strong correlation with the severity of AV shunt stenosis.

Arteriovenous Fistula Flow Dysfunction Surveillance: Early Detection Using Pulse Radar Sensor and Machine Learning Classification

“…Phono-angiography employs a digital stethoscope to record bruit sounds in the AVF. However, four locations are required in order to search for the most probable stenosis site [9,10]. Therefore, for the most part, detection sensitivity and specificity of the aforementioned methods do not meet the requirements for them to be reliable monitoring instruments.…”