Download This PaperDeep Learning Approach for Analysis of Artifacts in Heart Sound
14 Pages Posted: 17 Apr 2020
Date Written: April 16, 2020
Abstract
The cardiac system, also known as heart, is a significant part of the human body. Its gentle sound accompanies us throughout our lives, but if this sound gets corrupted or blocked by an abnormality, it is challenging for a human being to survive. In this way, the study of heart sounds and related abnormalities have a significant place in the area of biomedical engineering. Also, deep learning is one of the growing areas in the field of artificial intelli-gence which is leading the world to the invention of new smart devices that will be capable of making more accurate decisions. The thesis proposed deals with the artifact abnormality in heart sound and the objective is to classify the heart sounds into its three variants like normal, abnormal and artifact using deep learning. In the end, the performance of models is calculated with their accuracy and other performance metrics.
Keywords: Heart Sounds, deep learning, classification, hybrid neural networks, sequence models.
Suggested Citation: Suggested Citation