ECG signals used in the diagnosis of cardiovascular diseases are very important in terms of continuous recording and evaluation during the monitoring of these diseases, determination of appropriate diagnosis and treatment, and observation of possible complications. The most common disturbances among heart diseases are arising from arrhythmias. In this study, it was aimed to detect the cardiac arrhythmias APC and PVC automatically in the computer environment to provide convenience to the physician. In this context, ECG signals were first taken from the MIT-BIH Arrhythmia database and critical points P, Q, R, S, T on the signals were determined. After then, ANN was used for arrhythmia classification as APC, PVC and NSR. It was determined that the best result among the different ANN constructions was obtained with the MLPNN and the accuracy of the test was determined as 99.78% with 3-fold cross-validation and 99.89% with 10-fold cross-validation.
|Konular||Mühendislik, Elektrik ve Elektronik|
|Yayımlanma Tarihi||2020 Mart 8(1)|
|Bölüm||Araştırma Makalesi \ Research Makaleler|
: 21 Nisan 2019
|APA||AKIN, Z , BİLGİN, S . (2020). A NEW METHOD FOR THE AUTOMATIC DETECTION OF VENTRICULAR AND ATRIAL PREMATURE CONTRACTIONS. Mühendislik Bilimleri ve Tasarım Dergisi , 8 (1) , 165-174 . DOI: 10.21923/jesd.556486|