CD Skripsi
Deteksi Pola Sinyal Elektrokardiogram Irama Iskemia Miokardial Menggunakan Jaringan Syaraf Tiruan Berbasis Matlab Simulink
This study aims to design a computer program to detect myocardial ischemic
through ECG signal patterns and their accuracy. Myocardial ischemia is a heart
disorder due to narrowing of blood vessels in the heart wall. The method used is a
backpropagation ANN based on Simulink Matlab. Input data trained to recognize
target ECG signal patterns based on potential and time in the ST segment. The
optimal weight from training results used in testing program that has been made.
ANN outputs analyzed for potential depression or elevation to identify normal
heart or myocardial ischemia. The training results show that from several
architectures that have been tested, the optimal architecture is 1 hidden layer with
11 hidden units. These results are obtained from the Epoch parameters and the
value of MSE as well as the accuracy of each architecture. The learning process of
backpropagation ANN requires 8 epochs to achieve performance goals with an
MSE of 4.03×10-9. The system can recognize target patterns with training
accuracy of 99.82%. The results of testing identification system can detect
myocardial ischemic heart and normal heart disorders with an accuracy of 86.7%.
Based on the accuracy of the ANN program identification system, the detection of
myocardial ischemic rhythm ECG signal patterns using artificial neural networks
can be said to work well.
Keywords: Myocardial Ischemia, ECG, ANN Simulink Matlab
Tidak tersedia versi lain