Deep attention-based neural networks for explainable heart sound classification
Cardiovascular diseases are the leading cause of death and severely threaten human health in daily life. There have been dramatically increasing demands from both the clinical practice and the smart home application for monitoring the heart status of individuals suffering from chronic cardiovascular...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2022-09-01
|
| Series: | Machine Learning with Applications |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S266682702200038X |
| _version_ | 1798004290023849984 |
|---|---|
| author | Zhao Ren Kun Qian Fengquan Dong Zhenyu Dai Wolfgang Nejdl Yoshiharu Yamamoto Björn W. Schuller |
| author_facet | Zhao Ren Kun Qian Fengquan Dong Zhenyu Dai Wolfgang Nejdl Yoshiharu Yamamoto Björn W. Schuller |
| author_sort | Zhao Ren |
| collection | DOAJ |
| description | Cardiovascular diseases are the leading cause of death and severely threaten human health in daily life. There have been dramatically increasing demands from both the clinical practice and the smart home application for monitoring the heart status of individuals suffering from chronic cardiovascular diseases. However, experienced physicians who can perform efficient auscultation are still lacking in terms of number. Automatic heart sound classification leveraging the power of advanced signal processing and deep learning technologies has shown encouraging results. Nevertheless, a lack of explanation for deep neural networks is a limitation for the applications of automatic heart sound classification. To this end, we propose explaining deep neural networks for heart sound classification with an attention mechanism. We evaluate the proposed approach on the heart sounds shenzhen corpus. Our approach achieves an unweighted average recall of 51.2% for classifying three categories of heart sounds, i.e., normal, mild, and moderate/severe. The experimental results also demonstrate that the global attention pooling layer improves the performance of the learnt representations by estimating the contribution of each unit in high-level features. We further analyse the deep neural networks by visualising the attention tensors. |
| first_indexed | 2024-04-11T12:21:14Z |
| format | Article |
| id | doaj.art-3560530e27c2492cb3aadd3fb83eee6b |
| institution | Directory Open Access Journal |
| issn | 2666-8270 |
| language | English |
| last_indexed | 2024-04-11T12:21:14Z |
| publishDate | 2022-09-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Machine Learning with Applications |
| spelling | doaj.art-3560530e27c2492cb3aadd3fb83eee6b2022-12-22T04:24:05ZengElsevierMachine Learning with Applications2666-82702022-09-019100322Deep attention-based neural networks for explainable heart sound classificationZhao Ren0Kun Qian1Fengquan Dong2Zhenyu Dai3Wolfgang Nejdl4Yoshiharu Yamamoto5Björn W. Schuller6Chair of Embedded Intelligence for Health Care and Wellbeing, University of Augsburg, Augsburg 86159, Germany; L3S Research Center, Leibniz University Hannover, Hannover 30159, Germany; Corresponding author at: L3S Research Center, Leibniz University Hannover, Hannover 30159, Germany.School of Medical Technology, Beijing Institute of Technology, Beijing 100081, China; Corresponding author.University of Hong Kong – Shenzhen Hospital, Shenzhen 518009, China; Department of Cardiology, Shenzhen University General Hospital, Shenzhen 518055, ChinaDepartment of Cardiovascular, Wenzhou Medical University First Affiliated Hospital, Wenzhou 325035, ChinaL3S Research Center, Leibniz University Hannover, Hannover 30159, GermanyEducational Physiology Laboratory, Graduate School of Education, The University of Tokyo, Tokyo 113-0033, JapanChair of Embedded Intelligence for Health Care and Wellbeing, University of Augsburg, Augsburg 86159, Germany; GLAM – Group on Language, Audio & Music, Imperial College London, London SW7 2AZ, UKCardiovascular diseases are the leading cause of death and severely threaten human health in daily life. There have been dramatically increasing demands from both the clinical practice and the smart home application for monitoring the heart status of individuals suffering from chronic cardiovascular diseases. However, experienced physicians who can perform efficient auscultation are still lacking in terms of number. Automatic heart sound classification leveraging the power of advanced signal processing and deep learning technologies has shown encouraging results. Nevertheless, a lack of explanation for deep neural networks is a limitation for the applications of automatic heart sound classification. To this end, we propose explaining deep neural networks for heart sound classification with an attention mechanism. We evaluate the proposed approach on the heart sounds shenzhen corpus. Our approach achieves an unweighted average recall of 51.2% for classifying three categories of heart sounds, i.e., normal, mild, and moderate/severe. The experimental results also demonstrate that the global attention pooling layer improves the performance of the learnt representations by estimating the contribution of each unit in high-level features. We further analyse the deep neural networks by visualising the attention tensors.http://www.sciencedirect.com/science/article/pii/S266682702200038XComputer auditionHeart sound classificationSensor signal processingDigital health |
| spellingShingle | Zhao Ren Kun Qian Fengquan Dong Zhenyu Dai Wolfgang Nejdl Yoshiharu Yamamoto Björn W. Schuller Deep attention-based neural networks for explainable heart sound classification Machine Learning with Applications Computer audition Heart sound classification Sensor signal processing Digital health |
| title | Deep attention-based neural networks for explainable heart sound classification |
| title_full | Deep attention-based neural networks for explainable heart sound classification |
| title_fullStr | Deep attention-based neural networks for explainable heart sound classification |
| title_full_unstemmed | Deep attention-based neural networks for explainable heart sound classification |
| title_short | Deep attention-based neural networks for explainable heart sound classification |
| title_sort | deep attention based neural networks for explainable heart sound classification |
| topic | Computer audition Heart sound classification Sensor signal processing Digital health |
| url | http://www.sciencedirect.com/science/article/pii/S266682702200038X |
| work_keys_str_mv | AT zhaoren deepattentionbasedneuralnetworksforexplainableheartsoundclassification AT kunqian deepattentionbasedneuralnetworksforexplainableheartsoundclassification AT fengquandong deepattentionbasedneuralnetworksforexplainableheartsoundclassification AT zhenyudai deepattentionbasedneuralnetworksforexplainableheartsoundclassification AT wolfgangnejdl deepattentionbasedneuralnetworksforexplainableheartsoundclassification AT yoshiharuyamamoto deepattentionbasedneuralnetworksforexplainableheartsoundclassification AT bjornwschuller deepattentionbasedneuralnetworksforexplainableheartsoundclassification |