NUST Institutions Library Catalogue

Comparative Analysis of Machine Learning Techniques for Phonocardiogram-Based Classification of Cardiac Abnormalities / (Record no. 610839)

000 -LEADER
fixed length control field 02687nam a22001577a 4500
082 ## - DEWEY DECIMAL CLASSIFICATION NUMBER
Classification number 610
100 ## - MAIN ENTRY--PERSONAL NAME
Personal name Zaheer, Iqra
245 ## - TITLE STATEMENT
Title Comparative Analysis of Machine Learning Techniques for Phonocardiogram-Based Classification of Cardiac Abnormalities /
Statement of responsibility, etc. Iqra Zaheer
264 ## - PRODUCTION, PUBLICATION, DISTRIBUTION, MANUFACTURE, AND COPYRIGHT NOTICE
Place of production, publication, distribution, manufacture Islamabad :
Name of producer, publisher, distributor, manufacturer SMME- NUST;
Date of production, publication, distribution, manufacture, or copyright notice 2024.
300 ## - PHYSICAL DESCRIPTION
Extent 79p.
Other physical details Soft Copy
Dimensions 30cm
500 ## - GENERAL NOTE
General note Cardiovascular Diseases (CVDs) remain a leading cause of morbidity and mortality<br/>worldwide, necessitating early and accurate detection for effective disease management.<br/>This work employs advanced signal processing techniques in conjunction with machine<br/>learning methodologies to classify normal and particular cardiac conditions—Aortic<br/>Stenosis (AS), Mitral Regurgitation (MR), Mitral Stenosis (MS), and Mitral Valve<br/>Prolapse (MVP)—using phonocardiogram (PCG) signals. Preprocessing involved<br/>denoising using the Discrete Wavelet Transform (DWT) technique with the db8 wavelet<br/>and cA2 component, optimizing noise reduction while retaining valuable features for<br/>further analysis. Feature extraction was performed using Mel-Frequency Cepstral<br/>Coefficients (MFCC) and Mel Power Spectrogram, providing a robust and efficient<br/>representation of heart sounds. Two machine learning models—Deep Neural Network<br/>(DNN) and Convolutional Neural Network (CNN)—were used to assess the extracted<br/>features. With three hidden layers and 80% of the dataset used for training, the DNN model<br/>produced 90%±0.37 accuracy, 89% sensitivity, and 91% specificity. On the other hand, the<br/>CNN model, which consists of two fully connected layers and two convolutional layers<br/>with max pooling, performed by achieving 96%±0.38 accuracy, 95% sensitivity, and 95%<br/>specificity. These results underscore DNN’s enhanced capability in handling complex PCG<br/>data and reducing false negatives. This comprehensive study addresses multiple cardiac<br/>abnormalities, surpassing previous research that often focuses on a single condition or<br/>model. The findings highlight the potential of combining advanced signal processing with<br/>deep learning techniques to improve the timely and accurate identification of cardiac<br/>abnormalities. Future research will explore additional feature extraction methods and larger<br/>datasets to further enhance classification performance. This work significantly contributes<br/>to the field of biomedical engineering, offering a framework to improve patient outcomes<br/>through advanced diagnostic techniques.
650 ## - SUBJECT ADDED ENTRY--TOPICAL TERM
Topical term or geographic name entry element MS Biomedical Sciences
700 ## - ADDED ENTRY--PERSONAL NAME
Personal name Supervisor : Dr. Muhammad Asim Waris
856 ## - ELECTRONIC LOCATION AND ACCESS
Uniform Resource Identifier <a href="http://10.250.8.41:8080/xmlui/handle/123456789/45315">http://10.250.8.41:8080/xmlui/handle/123456789/45315</a>
942 ## - ADDED ENTRY ELEMENTS (KOHA)
Source of classification or shelving scheme
Koha item type Thesis
Holdings
Withdrawn status Permanent Location Current Location Shelving location Date acquired Full call number Barcode Koha item type
  School of Mechanical & Manufacturing Engineering (SMME) School of Mechanical & Manufacturing Engineering (SMME) E-Books 08/09/2024 610 SMME-TH-1042 Thesis
© 2023 Central Library, National University of Sciences and Technology. All Rights Reserved.