Phonocardiogram signals classification into normal heart sounds and heart murmur sounds

摘要

Heart disease is the biggest killer in the world, it is a serious public health problem facing the world today. This problem has not only attracted the attention of doctors and cardiologists, but also that of signal processing specialists who seek to effectively detect this disease by treating cardiac signals. This article proposes a heart sounds classification algorithm into two categories (Normal and Murmur) through a study of factors that influence this classification which leads to select those who give birth to optimal findings. The heart sounds classification was done by appealing the Dataset A and B of the Pascal (Pattern Analysis, Statistical Modelling and Computational Learning) Classifying Heart Sounds Challenge. This work achieves improved classification results in comparison to those achieved by the challenge finalists.