DETEKSI OTOMATIS KELAINAN JANTUNG MENGGUNAKAN HIDDEN MARKOV MODEL (HMM)

Jondri dan Achmad Rizal
Institut Teknologi Telkom, Bandung
jdn@ittelkom.ac.id dan arz@ittelkom.ac.id

ABSTRACT
The heart of patient having an heart attack must be controlled every time. Electrocardiogram signal (ECG) of patient is used as an indicator of the condition of patients heart. In this research, Hidden Markov Model (HMM) is used to distinguish Normal Sinus Rhythm (NSR) or Atrial Fibrillation (AF) from patient Electrocardiogram signal (ECG). Wavelet decomposition packet and c-means clustering is used to build feature vector and code book. The accuracy of this method has been of 95% (100% for NSR and 90% for AF). Segmentation of ECG signal is only performed for NSR.

Keywords: Electrocardiogram (ECG), Hidden Markov Model (HMM).

Paper ini dipublikasikan pada seminar KNS&I 2010 yang diselenggarakan oleh STIKOM Bali, 13 November 2010.

Paper lengkap dapat di unduh disini

Leave a Reply