Heart sounds are obtained by the beating heart and the resultant flow of blood through it (particularly, the turbulence created when the heart valves snap shut). A patient's heart problem is often inferred by the doctor after listening to the heartbeat sound. This technique is known as auscultation. Auscultation is the most difficult skill for a doctor, since this skill can be acquired only by experience. If these sounds can be separated, the doctors can make the diagnosis more accurately and easily and also the chances for finding a disease at an early stage increases. We propose a technological solution for the conventional auscultation which is based on source separation of heart sound (S1, S2, S3 and S4) using FastICA and its FPGA implementation. The heart sound mixtures are captured by placing the required number of sensors at appropriate positions in body. The signals can be used for observation. The sound mixtures to be separated into its sources were given as input to FastICA algorithm. The separated heartbeat sounds can be processed).
展开▼
