Cardiopulmonary sound separation method and system based on DAE-NMF-VMD

The invention discloses a cardiopulmonary sound separation method and system based on DAE-NMF-VMD. The method comprises the steps that the height expression of cardiopulmonary sound signals is extracted through DAE; grouping the heart and lung sounds according to different periods of the heart and lung sounds by using an NMF clustering method to realize separation of the mixed heart and lung sounds; and carrying out VMD denoising on the separated signals to obtain pure heart sound and pure lung sound. Compared with other deep learning-based methods, the method provided by the invention has the advantage that marked training data is not needed. It benefits from the periodic structure, providing separation performance superior to conventional methods. 本发明公开了一种基于DAE-NMF-VMD的心肺音分离方法及系统，包括：通过DAE提取心肺音信号的高度表达；使用NMF聚类方法根据心肺音不同周期对其进行分组实现混合心肺音分离；对分离后的信号进行VMD去噪得到纯净心音和纯净肺音。与其他基于深度学习方法相比，本发明方法的优点是不需要标记的训练数据。它得益于周期性结构，提供了优于传统方法的分离性能。