Screening for abnormal heart sounds and murmurs by implementing neural networks
Date
2007-03
Authors
Visagie, Claude
Journal Title
Journal ISSN
Volume Title
Publisher
Stellenbosch : University of Stellenbosch
Abstract
This thesis is concerned with the testing of an “auscultation jacket” as a means of recording
heart sounds and electrocardiography (ECG) data from patients. A classification system
based on Neural Networks, that is able to discriminate between normal and abnormal heart
sounds and murmurs, has also been developed . The classification system uses the recorded
data as training and testing data. This classification system is proposed to serve as an aid to
physicians in diagnosing patients with cardiac abnormalities. Seventeen normal participants
and 14 participants that suffer from valve-related heart disease have been recorded with the
jacket. The “auscultation jacket” shows great promise as a wearable health monitoring
aid for application in rural areas and in the telemedicine industry. The Neural Network
classification system is able to differentiate between normal and abnormal heart sounds
with a sensitivity of 85.7% and a specificity of 94.1%.
Description
Thesis (MScEng (Mechanical and Mechatronic Engineering))--University of Stellenbosch, 2007.
Keywords
Dissertations -- Mechanical engineering, Theses -- Mechanical engineering, Neural networks (Computer science), Heart -- Sounds, Auscultation, Stethoscopes, Cardiovascular system -- Diseases -- Diagnosis