José Millet

Atrial activity extraction for atrial fibrillation analysis using blind source separation

This contribution addresses the extraction of atrial activity (AA) from real electrocardiogram (ECG) recordings of atrial fibrillation (AF). We show the appropriateness of independent component analysis (ICA) to tackle this biomedical challenge when regarded as a blind source separation (BSS) problem. ICA is a statistical tool able to reconstruct the unobservable independent sources of bioelectric activity which generate, through instantaneous linear mixing, a measurable set of signals. The three key hypothesis that make ICA applicable in the present scenario are discussed and validated: 1) AA and ventricular activity (VA) are generated by sources of independent bioelectric activity; 2) AA and VA present non-Gaussian distributions; and 3) the generation of the surface ECG potentials from the cardioelectric sources can be regarded as a narrow-band linear propagation process. To empirically endorse these claims, an ICA algorithm is applied to recordings from seven patients with persistent AF. We demonstrate that the AA source can be identified using a kurtosis-based reordering of the separated signals followed by spectral analysis of the sub-Gaussian sources. In contrast to traditional methods, the proposed BSS-based approach is able to obtain a unified AA signal by exploiting the atrial information present in every ECG lead, which results in an increased robustness with respect to electrode selection and placement.

Spatiotemporal blind source separation approach to atrial activity estimation in atrial tachyarrhythmias

The analysis and characterization of atrial tachyarrhythmias requires, in a previous step, the extraction of the atrial activity (AA) free from ventricular activity and other artefacts. This contribution adopts the blind source separation (BSS) approach to AA estimation from multilead electrocardiograms (ECGs). Previously proposed BSS methods for AA extraction-e.g., independent component analysis (ICA)-exploit only the spatial diversity introduced by the multiple spatially-separated electrodes. However, AA typically shows certain degree of temporal correlation, with a narrowband spectrum featuring a main frequency peak around 3.5-9 Hz. Taking advantage of this observation, we put forward a novel two-step BSS-based technique which exploits both spatial and temporal information contained in the recorded ECG signals. The spatiotemporal BSS algorithm is validated on simulated and real ECGs from a significant number of atrial fibrillation (AF) and atrial flutter (AFL) episodes, and proves consistently superior to a spatial-only ICA method. In simulated ECGs, a new methodology for the synthetic generation of realistic AF episodes is proposed, which includes a judicious comparison between the known AA content and the estimated AA sources. Using this methodology, the ICA technique obtains correlation indexes of 0.751, whereas the proposed approach obtains a correlation of 0.830 and an error in the estimated signal reduced by a factor of 40%. In real ECG recordings, we propose to measure performance by the spectral concentration (SC) around the main frequency peak. The spatiotemporal algorithm outperforms the ICA method, obtaining a SC of 58.8% and 44.7%, respectively.

Comparison of atrial signal extraction algorithms in 12-lead ECGs with atrial fibrillation

Analysis of atrial rhythm is important in the treatment and management of patients with atrial fibrillation. Several algorithms exist for extracting the atrial signal from the electrocardiogram (ECG) in atrial fibrillation, but there are few reports on how well these techniques are able to recover the atrial signal. We assessed and compared three algorithms for extracting the atrial signal from the 12-lead ECG. The 12-lead ECGs of 30 patients in atrial fibrillation were analyzed. Atrial activity was extracted by three algorithms, Spatiotemporal QRST cancellation (STC), principal component analysis (PCA), and independent component analysis (ICA). The amplitude and frequency characteristics of the extracted atrial signals were compared between algorithms and against reference data. Mean (standard deviation) amplitude of QRST segments of V1 was 0.99 (0.54) mV, compared to 0.18 (0.11) mV (STC), 0.19 (0.13) mV (PCA), and 0.29 (0.22) mV (ICA). Hence, for all algorithms there were significant reductions in the amplitude of the ventricular activity compared with that in V1. Reference atrial signal amplitude in V1 was 0.18 (0.11) mV, compared to 0.17 (0.10) mV (STC), 0.12 (0.09) mV (PCA), and 0.18 (0.13) mV (ICA) in the extracted atrial signals. PCA tended to attenuate the atrial signal in these segments. There were no significant differences for any of the algorithms when comparing the amplitude of the reference atrial signal with that of the extracted atrial signals in segments in which ventricular activity had been removed. There were no significant differences between algorithms in the frequency characteristics of the extracted atrial signals. There were discrepancies in amplitude and frequency characteristics of the atrial signal in only a few cases resulting from notable residual ventricular activity for PCA and ICA algorithms. In conclusion, the extracted atrial signals from these algorithms exhibit very similar amplitude and frequency characteristics. Users of these algorithms should be observant of residual ventricular activities which can affect the analysis of the fibrillatory waveform in clinical practice.

Noninvasive Localization of Maximal Frequency Sites of Atrial Fibrillation by Body Surface Potential Mapping

Background—Ablation of high-frequency sources in patients with atrial fibrillation (AF) is an effective therapy to restore sinus rhythm. However, this strategy may be ineffective in patients without a significant dominant frequency (DF) gradient. The aim of this study was to investigate whether sites with high-frequency activity in human AF can be identified noninvasively, which should help intervention planning and therapy. Methods and Results—In 14 patients with a history of AF, 67-lead body surface recordings were simultaneously registered with 15 endocardial electrograms from both atria including the highest DF site, which was predetermined by atrial-wide real-time frequency electroanatomical mapping. Power spectra of surface leads and the body surface location of the highest DF site were compared with intracardiac information. Highest DFs found on specific sites of the torso showed a significant correlation with DFs found in the nearest atrium (&rgr;=0.96 for right atrium and &rgr;=0.92 for left atrium) and the DF gradient between them (&rgr;=0.93). The spatial distribution of power on the surface showed an inverse relationship between the frequencies versus the power spread area, consistent with localized fast sources as the AF mechanism with fibrillatory conduction elsewhere. Conclusions—Spectral analysis of body surface recordings during AF allows a noninvasive characterization of the global distribution of the atrial DFs and the identification of the atrium with the highest frequency, opening the possibility for improved noninvasive personalized diagnosis and treatment.

Body surface localization of left and right atrial high-frequency rotors in atrial fibrillation patients: A clinical-computational study

BACKGROUND Ablation is an effective therapy in patients with atrial fibrillation (AF) in which an electrical driver can be identified. OBJECTIVE The aim of this study was to present and discuss a novel and strictly noninvasive approach to map and identify atrial regions responsible for AF perpetuation. METHODS Surface potential recordings of 14 patients with AF were recorded using a 67-lead recording system. Singularity points (SPs) were identified in surface phase maps after band-pass filtering at the highest dominant frequency (HDF). Mathematical models of combined atria and torso were constructed and used to investigate the ability of surface phase maps to estimate rotor activity in the atrial wall. RESULTS The simulations show that surface SPs originate at atrial SPs, but not all atrial SPs are reflected at the surface. Stable SPs were found in AF signals during 8.3% ± 5.7% vs. 73.1% ± 16.8% of the time in unfiltered vs. HDF-filtered patient data, respectively (P < .01). The average duration of each rotational pattern was also lower in unfiltered than in HDF-filtered AF signals (160 ± 43 ms vs. 342 ± 138 ms; P < .01), resulting in 2.8 ± 0.7 rotations per rotor. Band-pass filtering reduced the apparent meandering of surface HDF rotors by reducing the effect of the atrial electrical activity occurring at different frequencies. Torso surface SPs representing HDF rotors during AF were reflected at specific areas corresponding to the fastest atrial location. CONCLUSION Phase analysis of surface potential signals after HDF filtering during AF shows reentrant drivers localized to either the left atrium or the right atrium, helping in localizing ablation targets.

Noninvasive mapping of human atrial fibrillation.

Introduction: Invasive high‐density mapping of atrial fibrillation (AF) has revealed different patterns of atrial activation ranging from single wavefronts to disorganized activation with multiple simultaneous wavefronts. Whether or not similar activation patterns can also be observed using body surface recordings is currently unknown, and was consequently evaluated in this study.

Estimation of atrial fibrillatory wave from single-lead atrial fibrillation electrocardiograms using principal component analysis concepts

A new method for the assessment of the atrial fibrillatory wave (AFW) from the ECG is presented. This methodology is suitable for signals registered from Holter systems, where the reduced number of leads is insufficient to exploit the spatial information of the ECG. The temporal dependence of the bio-electrical activity were exploited using principal component analysis. The main features of ventricular and atrial activity were extracted, and several basis signals for each subspace were determined. Hence, the estimated (AFW) are reconstructed exclusively from the basis signals that formed the atrial subspace. Its main advantage with respect to adaptive template subtraction techniques was its robustness to variations in the QRST morphology, which thus minimised QRST residua. The proposed approach was first validated using a database of simulated recordings with known atrial activity content. The estimated AFW was compared with the original AFW, obtaining correlation indices of 0.774±0.106. The suitability of this methodology for real recordings was also proven, though its application to a set of paroxysmal AF ECGs. In all cases, it was possible to detect the main frequency peak, which was between 4.6 Hz and 6.9 Hz for the patients under study.

Prognostic and therapeutic implications of dipyridamole stress cardiovascular magnetic resonance on the basis of the ischaemic cascade

Objective: To determine the prognostic and therapeutic implications of stress perfusion cardiovascular magnetic resonance (CMR) on the basis of the ischaemic cascade. Setting: Single centre study in a teaching hospital in Spain. Patients: Dipyridamole stress CMR was performed on 601 patients with ischaemic chest pain and known or suspected coronary artery disease. On the basis of the ischaemic cascade, patients were categorised in C1 (no evidence of ischaemia, n = 354), C2 (isolated perfusion deficit at stress first-pass perfusion imaging, n = 181) and C3 (simultaneous perfusion deficit and inducible wall motion abnormalities, n = 66). CMR-related revascularisation (n = 102, 17%) was defined as the procedure prompted by the CMR results and carried out within the next three months. Results: During a median follow-up of 553 days, 69 major adverse cardiac events (MACE), including 21 cardiac deaths, 14 non-fatal myocardial infarctions and 34 admissions for unstable angina with documented abnormal angiography were detected. In non-revascularised patients (n = 499), the MACE rate was 4% (14/340) in C1, 20% (26/128) in C2 and 39% (12/31) in C3 (adjusted p value = 0.004 vs C2 and <0.001 vs C1). CMR-related revascularisation had neutral effects in C2 (20% vs 19%, 1.1 (0.5 to 2.4), p = 0.7) but independently reduced the risk of MACE in C3 (39% vs 11%, 0.2 (0.1 to 0.7), p = 0.01). Conclusions: Dypiridamole stress CMR is able to stratify risk on the basis of the ischaemic cascade. A small group of patients with severe ischaemia—simultaneous perfusion deficit and inducible wall motion abnormalities—are at the highest risk and benefit most from MACE reduction due to revascularisation.

Modification of Ventricular Fibrillation Activation Patterns Induced by Local Stretching

Introduction: We hypothesize that local modifications in electrophysiological properties, when confined to zones of limited extent, induce few changes in the global activation process during ventricular fibrillation (VF). To test this hypothesis, we produced local electrophysiological modifications by stretching a circumscribed zone of the left ventricular wall in an experimental model of VF.

Noninvasive Assessment of the Complexity and Stationarity of the Atrial Wavefront Patterns During Atrial Fibrillation

A novel automated approach to quantitatively evaluate the degree of spatio-temporal organization in the atrial activity (AA) during atrial fibrillation (AF) from surface recordings, obtained from body surface potential maps (BSPM), is presented. AA organization is assessed by measuring the reflection of the spatial complexity and temporal stationarity of the wavefront patterns propagating inside the atria on the surface ECG, by means of principal component analysis (PCA). Complexity and stationarity are quantified through novel parameters describing the structure of the mixing matrices derived by the PCA of the different AA segments across the BSPM recording. A significant inverse correlation between complexity and stationarity is highlighted by this analysis. The discriminatory power of the parameters in identifying different groups in the set of patients under study is also analyzed. The obtained results present analogies with earlier invasive studies in terms of number of significant components necessary to describe 95% of the variance in the AA (four for more organized AF, and eight for more disorganized AF). These findings suggest that automated analysis of AF organization exploiting spatial diversity in surface recordings is indeed possible, potentially leading to an improvement in clinical decision making and AF treatment.