Simona Petrutiu

Atrial fibrillation and waveform characterization

The surface electrocardiogram (ECG) is a convenient, cost effective, and noninvasive tool for the study of atrial fibrillation (AF). It can be used to examine the hypothesized mechanisms of AF as well as to quantify and assess the effect of electrophysiological remodeling and the effectiveness of treatment on different types of AF. Time domain methods can be used to characterize the signal in the surface ECG. The authors described observations that can be obtained directly from the signal, such as the general characteristics of AF in the surface ECG and the ventricular response to AF. A discussion on commonly used methods to characterize atrial activity is also presented. These methods include cancellation techniques, vector analysis, and autocorrelation. Observations show that combining time and frequency domain methods provides a more thorough understanding of the characteristics of the atrial activity in the surface ECG. Whether the study of atrial activity in the surface ECG can be used to distinctively distinguish between different mechanisms of AF is not yet known, but further investigation can improve our understanding of these mechanisms and help with the management of this common arrhythmia

Analysis of the surface electrocardiogram to predict termination of atrial fibrillation: the 2004 computers in cardiology/physionet challenge

Paroxysmal atrial fibrillation (AF) is self-terminating by definition, but the mechanisms by which this occurs are not wall understood. Holter recordings were used to develop and test algorithms for distinguishing between AF segments that are Non-terminating (N), Terminating within a second (T), and Soon terminating (within a minute, S). From rhe training set, the peak frequency ranges (mean±SD) were 4.8-6.0 (5.3±0.4) Hz for T 4.7- 6.4 (5.2±0.6) Hz for S, 4.8-7.3 (6.5±0.8) Hz for N. In 8 of IO T recordings there was a decrease in peak frequency from the penultimate to the ultimare second and in 8 of 10 T recordings there was a decrease in peak power. The last second had a lower peak frequency for T when compared to S of the same patient in 9 of IO patients. Between Nand T we correctly classified 20 of 20 from the training set and 29 of 30 from the test set. Bemeen S and T we correctly classified 20 of 20 from the training set and 20 of 20 from the test set.

Manifestation of Left Atrial Events and Interatrial Frequency Gradients in the Surface Electrocardiogram During Atrial Fibrillation: Contributions from Posterior Leads

Background: In most patients, atrial fibrillation (AF) is initiated and maintained by pulmonary vein foci, but the relationship between left atrial (LA) events and the surface electrocardiogram (ECG) is largely unknown. We investigated whether LA events are reflected in the surface ECG and whether additional information can be obtained from recording posterior leads in patients with AF.

pPOP: Fast yet accurate parallel hierarchical clustering using partitioning

Hierarchical agglomerative clustering (HAC) is very useful but due to high CPU time and memory complexity its practical use is limited. Earlier, we proposed an efficient partitioning - partially overlapping partitioning (POP) - based on the fact that in HAC small and closely placed clusters are agglomerated initially, and only towards the end larger and distant clusters are agglomerated. Here, we present the parallel version of POP, pPOP. Theoretical analysis shows that, compared to the existing algorithms, pPOP achieves CPU time speed-up and memory scale-down of O(c) without compromising accuracy where c is the number of cells in the partition. A shared memory implementation shows that pPOP outperforms existing algorithms significantly.

High resolution electrocardiography optimised for recording pulses from electronic pacemakers: Evaluation of a new pacemaker sensing system

Pacemaker outputs are poorly recorded and displayed by the standard ECG. A new high resolution ECG system optimized for recording outputs from electronic pacemakers was evaluated. Three ECGs with different pulse settings were recorded in 42 patients with pacemakers at a sampling rate of 75 kHz using a new ECG acquisition module. Atrial and ventricular pulses were detected and the measured pulse widths and amplitudes were compared to the programmed values. For the atrium, the correlations between programmed and measured width and amplitude were 0.85 and 0.74, while for the ventricle they were 0.99 and 0.8. Leads II and V1 had the highest atrial amplitude, and leads V3, V4 and V5 had the highest ventricular amplitude. The new high resolution ECG pacemaker system dramatically improved the reproduction of pacemaker outputs, allowing accurate measurement of pulse duration for both atrial and ventricular pulses.

Efficient Parallel Hierarchical Clustering

Hierarchical agglomerative clustering (HAC) is a common clustering method that outputs a dendrogram showing all N levels of agglomerations where N is the number of objects in the data set. High time and memory complexities are some of the major bottlenecks in its application to real-world problems. In the literature parallel algorithms are proposed to overcome these limitations. But, as this paper shows, existing parallel HAC algorithms are inefficient due to ineffective partitioning of the data. We first show how HAC follows a rule where most agglomerations have very small dissimilarity and only a small portion towards the end have large dissimilarity. Partially overlapping partitioning (POP) exploits this principle and obtains efficient yet accurate HAC algorithms. The total number of dissimilarities is reduced by a factor close to the number of cells in the partition. We present pPOP, the parallel version of POP, that is implemented on a shared memory multiprocessor architecture. Extensive theoretical analysis and experimental results are presented and show that pPOP gives close to linear speedup and outperforms the existing parallel algorithms significantly both in CPU time and memory requirements.

The relationship between programmed pacemaker pulse amplitude and the surface electrocardiogram recorded amplitude: application of a new high-bandwidth electrocardiogram system

BACKGROUND Recording and displaying outputs from electronic pacemakers with electrocardiogram (ECG) recorders typically used in clinical practice have presented a number of technical limitations. We have recently reported on a new high-bandwidth ECG system and have shown that it is capable of reproducing accurate pulse amplitudes and durations from the body surface. In the present work, we have used our data to calculate a transform function between the programmed pacemaker output voltage and the amplitude on the body surface. METHODS We recorded 3 high-bandwidth (75,000 samples per second) ECGs from each of 100 pacemaker patients at 3 different programmed outputs. Each pacemaker pulse was isolated using the criterion standard annotations, and the pulses were transformed from the 8 independent leads to an XYZ vector using the Dower transform. The magnitude of the vector was calculated. Linear regression techniques were used to learn a transfer function over the records of the first 50 patients. These results were tested against the second 50 patients. RESULTS The measured pacemaker pulse vector magnitude has a linear relationship to the programmed pacemaker amplitude on a per-patient basis for most of the patients in the training database. The linear transform models were tested against the testing set with an R(2) metric of 0.38 for the atrial pulses and 0.54 for the right ventricular pulses. CONCLUSION Understanding the relationship between the generated pacemaker pulses and the measurements at the body surface will help drive specifications for pacemaker pulse detection among the various device manufactures.