FUNCTIONAL AND MORPHOLOGICAL ANALYSIS OF BLOOD ACCESS BASED ON WAVELET TRANSFORM OF SHUNT MURMURS

Abstract

Maintaining blood access in a good condition is crucial for patients undergoing hemodialysis. At present, blood access function is generally assessed under stethoscopy by dialysis staff in clinical settings. When performed carefully, stethoscopy is capable of correctly assessing blood access function 70–80% of the time, but quantifiability and objectivity is lacking. Furthermore, results depend on staff experience and judgment, and ascertaining chronological changes is difficult as data cannot be stored. We have therefore been investigating noninvasive, quantitative and objective methods by which dialysis staff can routinely assess the function and morphology of blood access based on wavelet transformation. After constructing a shunt murmur measurement system using a commercially available heart sound sensor to obtain the frequency response of shunt murmurs and performing analysis based on wavelet transformation, we confirmed that the state of blood access can be classified into 4 groups. As a clinical application of this technique, blood access function was assessed before and after percutaneous transluminal angioplasty (PTA) or use of an artificial shunt. Furthermore, pulsations caused by sending pulsatile flow to a constricted blood vessel model were captured using the heart sound sensor, and signals were analyzed by wavelet transformation. The results broadly matched the clinical data.