Klasifikasi Penyakit Jantung Pada Suara Detak Jantung Menggunakan Hjorth Descriptor

Authors

  • ririn amrina . Politeknik Caltex Riau

Abstract

An early recognition tool for normal and abnormal heart sounds is needed to assist medical personnel in diagnosing abnormalities in the heart. Basic Health Research data shows that the death rate from non-communicable diseases in Indonesia has increased to 59.5% with deaths from heart disease occupying the seventh position of causes of death in various countries in the world. The design aims to identify two types of heart sounds, namely normal and abnormal (murmur), this system uses the Hjorth Descriptor feature extraction process. change activity values and variations. The classification process is carried out using MLP, MLP is able to analyze problems and perform pattern classification and system modeling in retrieval and determining features. The process of making this TA uses MATLAB software tools which will test the database that has been obtained which will later be designed on the GUI (Grapichal User Interface). In this final project the classification results obtained can be used with the possibility of a more accurate truth, in the results of this study, the highest level of accuracy is 100 % and the sensitivity value is 100 % and the specificity value is 100 % Keywords: Murmur, Hjorth Descriptor, MATLAB.

Published

2021-03-12

Issue

Section

Artikel