Note on a Markov chain approach to cardiac arrhythmia classification

✍ Scribed by Chelsea C. White III

• Publisher: Elsevier Science
• Year: 1976
• Tongue: English
• Weight: 225 KB
• Volume: 9
• Category: Article
• ISSN: 0010-4809
• DOI: 10.1016/0010-4809(76)90009-4

✦ Synopsis

A heart-beat interval, Markov chain approach to cardiac arrhythmia classification is reexamined, and an alternative problem formulation presented. The approach developed models heart-beat interval and underlying disorder as a partially observed, twodimensional Markov chain, where the scalar element modeling underlying disorder is assumed to always make transition back into the same state. The R-R wave interval is assumed perfectly observed and the underlying disorder completely unobserved. A recursive procedure for updating the probability density of an underlying disorder conditioned on all past data is stated. Selection of the classification which maximizes the probability of being the underlying disorder, given the observed sequence, is then easily determined.