A novel multi-branch architecture for state of the art robust detection of pathological phonocardiograms

Heart auscultation is an inexpensive and fundamental technique to effectively diagnose cardiovascular disease. However, due to relatively high human error rates even when auscultation is performed by an experienced physician, and due to the not universal availability of qualified personnel, e.g. in developing countries, many efforts are made worldwide to propose computational tools for detecting abnormalities in heart sounds. The large heterogeneity of achievable data quality and devices, the variety of possible heart pathologies, and a generally poor signal-to-noise ratio make this problem very challenging. We present an accurate classification strategy for diagnosing heart sounds based on (1) automatic heart phase segmentation, (2) state-of-the art filters drawn from the field of speech synthesis (mel-frequency cepstral representation) and (3) an ad hoc multi-branch, multi-instance artificial neural network based on convolutional layers and fully connected neuronal ensembles which separately learns from each heart phase hence implicitly leveraging their different physiological significance. We demonstrate that it is possible to train our architecture to reach very high performances, e.g. an area under the curve of 0.87 or a sensitivity of 0.97. Our machine-learning-based tool could be employed for heartsound classification, especially as a screening tool in a variety of situations including telemedicine applications. This article is part of the theme issue 'Advanced computation in cardiovascular physiology: new challenges and opportunities'.