Enhanced heart sound anomaly detection via WCOS: a semi-supervised framework integrating wavelet, autoencoder and SVM
Anomaly detection is a typical binary classification problem under the condition of unbalanced samples, which has been widely used in various fields of data mining. For example, it can help detect heart murmurs when the heart is structurally abnormal, to tell if a newborn has congenital heart diseas...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2025-01-01
|
| Series: | Frontiers in Neuroinformatics |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fninf.2025.1530047/full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832582901116436480 |
|---|---|
| author | Peipei Zeng Shuimiao Kang Shuimiao Kang Fan Fan Jiyuan Liu Jiyuan Liu |
| author_facet | Peipei Zeng Shuimiao Kang Shuimiao Kang Fan Fan Jiyuan Liu Jiyuan Liu |
| author_sort | Peipei Zeng |
| collection | DOAJ |
| description | Anomaly detection is a typical binary classification problem under the condition of unbalanced samples, which has been widely used in various fields of data mining. For example, it can help detect heart murmurs when the heart is structurally abnormal, to tell if a newborn has congenital heart disease. Due to the low time and high efficiency, most work focuses on the semi- supervised anomaly detection method. However, the anomaly detection effect of this method is not high because of massive data with uneven samples and different noise. To improve the accuracy of anomaly detection under unbalanced sample conditions, we propose a new semi-supervised anomaly detection method (WCOS) based on semi-supervised clustering, which combines wavelet reconstruction, convolutional autoencoder, and one classification support vector machine. In this way, we can not only distinguish a small proportion of abnormal heart sounds in the huge data scale but also filter the noise through the noise reduction network, thus significantly improving the detection accuracy. In addition, we evaluated our method using real datasets. When the noise of sigma = 0.5, the AUC standard deviation of the WR-CAE-OCSVM is 19.2, 54.1, and 29.8% lower than that of WR-OCSVM, CAE-OCSVM and OCSVM, respectively. The results confirmed the higher accuracy of anomaly detection in WCOS compared to other state-of-the-art methods. |
| format | Article |
| id | doaj-art-59e646350ce746aba235b5d7fd35df0d |
| institution | Kabale University |
| issn | 1662-5196 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Neuroinformatics |
| spelling | doaj-art-59e646350ce746aba235b5d7fd35df0d2025-01-29T06:45:49ZengFrontiers Media S.A.Frontiers in Neuroinformatics1662-51962025-01-011910.3389/fninf.2025.15300471530047Enhanced heart sound anomaly detection via WCOS: a semi-supervised framework integrating wavelet, autoencoder and SVMPeipei Zeng0Shuimiao Kang1Shuimiao Kang2Fan Fan3Jiyuan Liu4Jiyuan Liu5Civil Aviation University of China Engineering Technology Training Center, Civil Aviation University of China, Tianjin, ChinaAviation Ground Special Equipment Research Base of Civil Aviation Administration of China, Civil Aviation University of China, Tianjin, ChinaCollege of Safety Science and Engineering, Civil Aviation University of China, Tianjin, ChinaPediatric Cardiac Center, Beijing Children’s Hospital, Capital Medical University, Beijing, ChinaAviation Ground Special Equipment Research Base of Civil Aviation Administration of China, Civil Aviation University of China, Tianjin, ChinaCollege of Aeronautical Engineering, Civil Aviation University of China, Tianjin, ChinaAnomaly detection is a typical binary classification problem under the condition of unbalanced samples, which has been widely used in various fields of data mining. For example, it can help detect heart murmurs when the heart is structurally abnormal, to tell if a newborn has congenital heart disease. Due to the low time and high efficiency, most work focuses on the semi- supervised anomaly detection method. However, the anomaly detection effect of this method is not high because of massive data with uneven samples and different noise. To improve the accuracy of anomaly detection under unbalanced sample conditions, we propose a new semi-supervised anomaly detection method (WCOS) based on semi-supervised clustering, which combines wavelet reconstruction, convolutional autoencoder, and one classification support vector machine. In this way, we can not only distinguish a small proportion of abnormal heart sounds in the huge data scale but also filter the noise through the noise reduction network, thus significantly improving the detection accuracy. In addition, we evaluated our method using real datasets. When the noise of sigma = 0.5, the AUC standard deviation of the WR-CAE-OCSVM is 19.2, 54.1, and 29.8% lower than that of WR-OCSVM, CAE-OCSVM and OCSVM, respectively. The results confirmed the higher accuracy of anomaly detection in WCOS compared to other state-of-the-art methods.https://www.frontiersin.org/articles/10.3389/fninf.2025.1530047/fullheart sound detectionsemi-supervised anomaly detectionsample imbalanceconvolutional autoencoderone classification support vector machine |
| spellingShingle | Peipei Zeng Shuimiao Kang Shuimiao Kang Fan Fan Jiyuan Liu Jiyuan Liu Enhanced heart sound anomaly detection via WCOS: a semi-supervised framework integrating wavelet, autoencoder and SVM Frontiers in Neuroinformatics heart sound detection semi-supervised anomaly detection sample imbalance convolutional autoencoder one classification support vector machine |
| title | Enhanced heart sound anomaly detection via WCOS: a semi-supervised framework integrating wavelet, autoencoder and SVM |
| title_full | Enhanced heart sound anomaly detection via WCOS: a semi-supervised framework integrating wavelet, autoencoder and SVM |
| title_fullStr | Enhanced heart sound anomaly detection via WCOS: a semi-supervised framework integrating wavelet, autoencoder and SVM |
| title_full_unstemmed | Enhanced heart sound anomaly detection via WCOS: a semi-supervised framework integrating wavelet, autoencoder and SVM |
| title_short | Enhanced heart sound anomaly detection via WCOS: a semi-supervised framework integrating wavelet, autoencoder and SVM |
| title_sort | enhanced heart sound anomaly detection via wcos a semi supervised framework integrating wavelet autoencoder and svm |
| topic | heart sound detection semi-supervised anomaly detection sample imbalance convolutional autoencoder one classification support vector machine |
| url | https://www.frontiersin.org/articles/10.3389/fninf.2025.1530047/full |
| work_keys_str_mv | AT peipeizeng enhancedheartsoundanomalydetectionviawcosasemisupervisedframeworkintegratingwaveletautoencoderandsvm AT shuimiaokang enhancedheartsoundanomalydetectionviawcosasemisupervisedframeworkintegratingwaveletautoencoderandsvm AT shuimiaokang enhancedheartsoundanomalydetectionviawcosasemisupervisedframeworkintegratingwaveletautoencoderandsvm AT fanfan enhancedheartsoundanomalydetectionviawcosasemisupervisedframeworkintegratingwaveletautoencoderandsvm AT jiyuanliu enhancedheartsoundanomalydetectionviawcosasemisupervisedframeworkintegratingwaveletautoencoderandsvm AT jiyuanliu enhancedheartsoundanomalydetectionviawcosasemisupervisedframeworkintegratingwaveletautoencoderandsvm |