Phonocardiogram signal analysis using adaptive line enhancer methods on Mixed Signal Processor

This paper presents the acquisition of heart sound signals and its separation from lung sounds based on the adaptive filtering technique. After signal acquisition, normal linear filtering techniques is not capable separated the two sounds, as both the sounds overlap in the spatial as well as temporal domains. Because of the signal tracking and noise elimination capability of adaptive line enhancer (ALE) it can be used for denoising the heart sounds. Here two methods of ALE have been discussed; using least mean square (LMS) and recursive least square (RLS). A comparative analysis has been carried out in terms of their power spectrum density (PSD) and percentage of heart sound recovery for their evaluation and finally both the algorithms have been implemented on MSP430 platform for designing a standalone diagnostic system to assist the cardiologists in their diagnosis. The scope of this study includes incorporation of the different diagnostic features into a single system for properly classifying the abnormal cardiac conditions.