Taking into account that heart auscultation remains the dominant method for heart examination in the small health centers of the rural areas and generally in primary healthcare set-ups, the enhancement of this technique would aid significantly in the diagnosis of heart diseases. In this context, a more systematic, orderly and intelligent method compares to traditional ones, was initially proposed in extracting the anomaly characteristic of heart sounds. The heart sounds segmentation algorithm was established based on the heart sounds extracted by medical sensor. A further performance analysis is presented. The experimental results indicate that the performance of this method is excellent. It is much more efficient and accurate than the ordinary methods.