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

Abstract

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’.

Deep Learning for Heart Sound Analysis: A Literature Review

“…In the reviewed literature, most studies use CNN models with 2 to 34 convolutional layers [9, 10, 11, 13, 14, 15, 18, 21, 22, 23, 24, 25, 26, 27, 28, 31, 32, 33, 34, 35, 37, 38, 40, 41, 42, 47, 48, 50, 51, 52, 55, 58, 59, 60, 61, 68, 69, 71, 72, 73, 84], which are usually equipped with rectified linear units, batch normalization, dropout and pooling components, and some of the layers are linked by residual connections. Wang et al [76] test 10 different CNN models including GoogleNet, SqueezeNet, DarkNet19, ModileNetv2, Inception-ResNetv2, DenseNet201, Inceptionv3, ResNet101, NasNet-Large, and Xception to compare the performances.…”

“…Since detecting cardiac murmurs is a relatively more straightforward task compared to identifying specific diseases, there has been a notable accumulation of high-quality annotated data in recent years, which has been made available to the public [20, 94]. The abundance of accessible data has led to a surge in DL models research on cardiac murmurs detection [21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56]. Although these models can only discern the presence of cardiac murmurs and cannot provide definitive diagnoses, they still play a crucial role in community-based disease screening.…”

“…At last, 71 original articles were included. These studies can be broadly categorized into several groups: methods (15 papers, including heart sound segmentation [6, 7, 8, 9, 10, 11, 12, 13], noise cancellation [14, 15, 16], algorithm development [17, 18, 19], and database development [20]), cardiac murmurs detection (36 papers [21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56]), valvular heart disease (6 papers [57, 58, 59, 60, 61, 62]), congenital heart disease (4 papers [63, 64, 65, 66]), heart failure (4 papers [67, 68, 69, 70]), coronary artery disease (2 papers [71, 72]), rheumatic heart disease (2 papers [73, 74]), and extracardiac applications (2 papers [75, 76]).…”

“…The abundance of accessible data has led to a surge in DL models research on cardiac murmurs detection[21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56]. Although these models can only discern the presence of cardiac murmurs and cannot provide definitive diagnoses, they still play a crucial role in community-based disease screening.…”

Advanced computation in cardiovascular physiology: new challenges and opportunities

“…Two further developments of AI methods for cardiology research are included in the theme issue. Duggento et al [5] show a novel multi-branch convolutional neural network architecture and pre-processing pipeline for a robust detection of pathological phonocardiograms. The study overcomes the high interpretation variability in human decision-making regarding abnormalities detected through heart sounds auscultation.…”