No title
Circadian variation in atrial fibrillation (AF) frequency is explored in this paper by employing recent advances in signal processing. Once the AF frequency has been estimated and tracked by a hidden Markov model approach, the resulting trend is analyzed for the purpose of detecting and characterizing the presence of circadian variation. With cosinor analysis, the results show that the short-term
