An accurate valvular heart disorders detection model based on a new dual symmetric tree pattern using stethoscope sounds

• Published in: Computers in biology and medicine Vol. 146; p. 105599
• Main Authors: Barua, Prabal Datta, Karasu, Mehdi, Kobat, Mehmet Ali, Balık, Yunus, Kivrak, Tarık, Baygin, Mehmet, Dogan, Sengul, Demir, Fahrettin Burak, Tuncer, Turker, Tan, Ru-San, Acharya, U. Rajendra
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01.07.2022; Elsevier Limited
• Subjects: Accuracy; Acoustics; Artificial intelligence; Cardiologic disorders detection; Cardiovascular disease; Classification; Computer applications; Coronary artery disease; Datasets; Deep learning; Diagnosis; Discrete Wavelet Transform; Dual symmetric tree pattern; Echocardiography; Feature extraction; Heart; Heart diseases; Internal Medicine; Iterative methods; Machine learning; Morbidity; Multilevel; Neural networks; Other; Sound; Stethoscope sound classification; Support vector machines; Wavelet transforms; Stethoscope sound classification; Cardiologic disorders detection; Dual symmetric tree pattern; Machine learning
• DOI: 10.1016/j.compbiomed.2022.105599
• ISSN: 0010-4825; 1879-0534

Valvular heart disease (VHD) is an important cause of morbidity and mortality. Echocardiography is the reference standard for VHD diagnosis but is not universally accessible. Manual cardiac auscultation is inadequate for screening VHD. Many machine learning models using heart sounds acquired with an electronic stethoscope may improve the accuracy of VHD diagnosis. We aimed to develop an accurate sound classification model for VHD diagnosis. A new large stethoscope sound dataset containing 10,366 heart sounds divided into ten categories (nine VHD and one healthy) were prospectively collected. We developed a handcrafted learning model that comprised multilevel feature extraction based on a dual symmetric tree pattern (DSTP) and multilevel discrete wavelet transform (DWT), feature selection, and classification. The multilevel DWT was used to create subbands to extract features at both high and low levels. Then, iterative neighborhood component analysis was used to select the most discriminative 512 features from among the extracted features in the generated feature vector. In the classification phase, a support vector machine (SVM) was used with 10-fold cross-validation (CV) and leave-one-subject-out (LOSO) CV. Our proposed DSTP-based model attained 99.58% and 99.84% classification accuracies using SVM classifier with 10-fold CV and LOSO CV, respectively. The presented DSTP-based classification model attained excellent multiclass classification performance on a large prospective heart sound dataset at a low computational cost. •A large stethoscope sound dataset was collected with 10 categories.•New textural feature extractor (DSTP) was proposed.•We developed a new hand-modeled sound classification.•10-fold CV and LOSO CV have been used to get robust results.•Our model outperformed.