BACKGROUNDS: Atrial fibrillation (AF) is the most common cardiac arrhythmia, becomes more prevalent with age and is associated with an increased long-term risk of stroke, heart failure, and all-cause mortality. Over the past decade, catheter ablation has emerged as a promising approaches for treating AF. Patients with stroke occasionally are firstly diagnosed with AF during admission. There is need to detect silent or asymptomatic AF episodes among AF patients with antiarrhythmic drug therapy, or patients after radiofrequency catheter ablation (RFCA), or patients with stroke. However, ECG or Holter monitoring have a limitation to detect AF episodes. OBJECTIVS: We evaluated 24hr Holter monitoring for detection of silent or asymptomatic atrial fibrillation episodes using new image analysis of pointcare plot. DESIGN: We collected data of 48hr Hoter monitoring for 33 patients during 1 year and evaluated sensitivity and specificity of new image analysis. According to Holter monitoring, 14 patients were diagnosed as PAF and 19 patients were diagnosed as normal sinus rhythm. The standard measures of heart rate and its variability in the time domain are considered linear measures. These include the average R-R interval (Mean RR) and its standard deviation (SDRR). Another approach is to use the differences in successive R-R intervals (successive deltas) and estimate the standard deviation (SDSD) of these deltas. The pointcare plot was already known method to evaluate R-R interval, but we used new analyzing technique of pointcare plot. RESULTS: New image analysis of pointcare plot correctly diagnosed AF (sensitivity) in 72.2% and sinus rhythm (specificity) in 92.8%. We did a follow-up for patients who were diagnosed as normal sinus rhythm using Holter monitoring but were diagnosed as PAF using new image analysis of pointcare plot. Finally we found silent AF episodes for 4 patients who were diagnosed as normal sinus rhythm with Holter monitoring. All false positive cases have relatively high proportion of noise at Holter monitoring. CONCLUSIONS: New image analysis of pointcare plot can detect silent AF episodes. This new analyzing technique will be used for AF patients who undertaken RFCA or who maintain sinus rhythm with antiarrhythmic drugs.