Optimal Multi-Stage Arrhythmia Classification Approach

Arrhythmia constitutes a problem with the rate or rhythm of the heartbeat, and an early diagnosis is essential for the timely inception of successful treatment. We have jointly optimized the entire multi-stage arrhythmia classification scheme based on 12-lead surface ECGs that attains the accuracy performance level of professional cardiologists. The new approach is comprised of a three-step noise reduction stage, a novel feature extraction method and an optimal classification model with finely tuned hyperparameters. We carried out an exhaustive study comparing thousands of competing classification algorithms that were trained on our proprietary, large and expertly labeled dataset consisting of 12-lead ECGs from 40,258 patients with four arrhythmia classes: atrial fibrillation, general supraventricular tachycardia, sinus bradycardia and sinus rhythm including sinus irregularity rhythm. Our results show that the optimal approach consisted of Low Band Pass filter, Robust LOESS, Non Local Means smoothing, a proprietary feature extraction method based on percentiles of the empirical distribution of ratios of interval lengths and magnitudes of peaks and valleys, and Extreme Gradient Boosting Tree classifier, achieved an F1-Score of 0.988 on patients without additional cardiac conditions. The same noise reduction and feature extraction methods combined with Gradient Boosting Tree classifier achieved an F1-Score of 0.97 on patients with additional cardiac conditions. Our method achieved the highest classification accuracy (average 10-fold cross-validation F1-Score of 0.992) using an external validation data, MIT-BIH arrhythmia database. The proposed optimal multi-stage arrhythmia classification approach can dramatically benefit automatic ECG data analysis by providing cardiologist level accuracy and robust compatibility with various ECG data sources.

Non-Contact, Simple Neonatal Monitoring by Photoplethysmography

This paper presents non-contact vital sign monitoring in neonates, based on image processing, where a standard color camera captures the plethysmographic signal and the heart and breathing rates are processed and estimated online. It is important that the measurements are taken in a non-invasive manner, which is imperceptible to the patient. Currently, many methods have been proposed for non-contact measurement. However, to the best of the authors’ knowledge, it has not been possible to identify methods with low computational costs and a high tolerance to artifacts. With the aim of improving contactless measurement results, the proposed method based on the computer vision technique is enhanced to overcome the mentioned drawbacks. The camera is attached to an incubator in the Neonatal Intensive Care Unit and a single area in the neonate’s diaphragm is monitored. Several factors are considered in the stages of image acquisition, as well as in the plethysmographic signal formation, pre-filtering and filtering. The pre-filter step uses numerical analysis techniques to reduce the signal offset. The proposed method decouples the breath rate from the frequency of sinus arrhythmia. This separation makes it possible to analyze independently any cardiac and respiratory dysrhythmias. Nine newborns were monitored with our proposed method. A Bland-Altman analysis of the data shows a close correlation of the heart rates measured with the two approaches (correlation coefficient of 0.94 for heart rate (HR) and 0.86 for breath rate (BR)) with an uncertainty of 4.2 bpm for HR and 4.9 for BR (k = 1). The comparison of our method and another non-contact method considered as a standard independent component analysis (ICA) showed lower central processing unit (CPU) usage for our method (75% less CPU usage).

Association between paroxysmal atrial fibrillation and the left atrial appendage ejection fraction during sinus rhythm in the acute stage of stroke: a transesophageal echocardiographic study

BACKGROUND

This study aimed to investigate whether left atrial appendage (LAA) dysfunction evaluated by transesophageal echocardiography (TEE) during sinus rhythm is predictable of paroxysmal atrial fibrillation (PAF) as an embolic source in the acute stage of stroke.

METHODS AND RESULTS

We measured and analyzed LAA flow velocity (LAA-FV) and LAA ejection fraction (LAA-EF) in 300 acute ischemic stroke patients by TEE. We divided the acute ischemic stroke patients into 3 groups. The atrial fibrillation (AF) group (n=58) comprised patients whose TEE was performed during AF rhythm. The PAF group (n=42) comprised patients with a history of AF but with normal sinus rhythm when TEE was performed. The normal sinus (sinus) group (n=200) did not have any history of AF. We found that mean LAA-FV and LAA-EF values in the PAF group were significantly lower than those in the sinus group (P<.001). The diagnostic accuracy of LAA-FV for the diagnosis of PAF calculated as the area under receiver operating characteristic curves was low (.582, 95% confidence interval [CI]=.498-.665) but that of LAA-EF was modest (.721, 95% CI=.653-.789), with an optimal cutoff point of 49.1%.

CONCLUSIONS

LAA dysfunction as determined by TEE (LAA-EF<49.1%) in the acute stage of stroke is predictive of PAF with moderate accuracy. Long-term electrocardiographic monitoring is recommended for cryptogenic stroke patients with LAA dysfunction.

Relation of prolonged tissue Doppler imaging-derived atrial conduction time to atrial arrhythmia in adult patients with congenital heart disease

Atrial arrhythmia (AA) is common in adult patients with congenital heart disease (CHD). To enable the prevention of AA or its complications, timely identification of adult patients with CHD at risk of AA is crucial. Long total atrial activation times have been related to AA. Tissue Doppler imaging (TDI) permits noninvasive evaluation of the total atrial conduction time (PA-TDI duration). The present study evaluated the association between the PA-TDI duration and the development of AA in adult patients with CHD. A total of 223 adult patients with CHD were followed up for the occurrence of AA after PA-TDI duration assessment. The PA-TDI duration was defined as the interval from the onset of the P wave on the electrocardiogram to the peak of the A' wave at the lateral atrial wall on TDI tracings. Among the various clinical and echocardiographic parameters, the association between the PA-TDI duration and AA occurrence was investigated. The median follow-up was 39 months (interquartile range 21 to 57). A PA-TDI duration of ≥126 ms was associated with AA during follow-up (log-rank, p <0.001). On multivariate analysis, a PA-TDI duration >126 ms (hazard ratio 2.25, 95% confidence interval 1.21 to 4.19) and history of AA (hazard ratio 4.89, 95% confidence interval 2.75 to 8.71) were independently associated with the occurrence of AA. In conclusion, PA-TDI duration and a history of AA were independently associated with the occurrence of AA in adult patients with CHD. The PA-TDI duration is a useful tool to identify patients with CHD at risk of AA during follow-up.

Extreme respiratory sinus arrhythmia enables overwintering black bear survival--physiological insights and applications to human medicine

American black bears survive winter months without food and water while in a mildly hypothermic, hypometabolic, and inactive state, yet they appear to be able to return to near-normal systemic function within minutes of arousal. This study's goal was to characterize the cardiovascular performance of overwintering black bears and elicit the underlying mechanisms enabling survival. Mid-winter cardiac electrophysiology was assessed in four wild black bears using implanted data recorders. Paired data from early and late winter were collected from 37 wild bears, which were anesthetized and temporarily removed from their dens to record cardiac electrophysiological parameters (12-lead electrocardiograms) and cardiac dimensional changes (echocardiography). Left ventricular thickness, primary cardiac electrophysiological parameters, and cardiovascular response to threats ("fight or flight" response) were preserved throughout winter. Dramatic respiratory sinus arrhythmias were recorded (cardiac cycle length variations up to 865%) with long sinus pauses between breaths (up to 13 s). The accelerated heart rate during breathing efficiently transports oxygen, with the heart "resting" between breaths to minimize energy usage. This adaptive cardiac physiology may have broad implications for human medicine.

Left Atrial Appendage Function in Patients with Systemic Embolism in Spite of in Sinus Rhythm

OBJECTIVE

We sought to investigate whether left atrial (LA) appendage (LAA) function was impaired in patients with systemic embolism in sinus rhythm.

METHODS

Transesophageal echocardiography was performed in 7 patients within 5 days after embolism (early group) and in 32 patients, more than 5 days after embolism (late group). We searched intracardiac thrombus and determined LAA area and LAA flow velocity, and the grade of spontaneous echocontrast.

RESULTS

LA thrombus could be detected in 2 of the 39 patients only in early group (P = .028). The patients of early group, as compared with late group, had lower LAA flow velocity (28 +/- 16 vs 60 +/- 26 cm/s, P = .007) and higher grade of spontaneous echocontrast score (1.14 +/- 1.46 vs 0.25 +/- 0.62, P = .013). LA dimension and LAA area were not statistically different.

CONCLUSION

LAA function may be impaired early after embolism even in sinus rhythm.

[Recovery of left atrial function after cardioversion of atrial fibrillation: the role of some clinical and echocardiographical factors]

Sinus rhythm was restored in 102 patients with atrial fibrillation. Cardioversion was spontaneous (n=6), pharmacological (n=35), and electrical - early (n=46) or postponed because of necessity of full anticoagulation (n=15). Irrespective of method of cardioversion all patients had signs of atrial stunning. Doppler echocardiography performed within 24 hours and in 3, 7 and 14 days after cardioversion showed that rate of recovery of mechanical left atrial function had pronounced individual variations. In patients with spontaneous termination of atrial fibrillation normalization of left atrial contractile activity occurred in about 3 days, after pharmacological cardioversion - in 1 week (in 91.4% of cases). After early and postponed electrical cardioversion mechanical left atrial function recovered in 2 weeks in 97.8 and 73.3% of patients, respectively. Most important impact on this process produced duration of the last attack of atrial fibrillation and left atrial dimensions.

Early pacemaker twiddler syndrome

Twiddler's syndrome is a well-known complication of pacemaker treatment. It was first described by Bayliss et al. when a patient manipulated and rotated the pulse generator in the pocket so many turns that it resulted in lead dislodgment, diaphragmatic stimulation and loss of capture. In this case report we present a patient who managed to rotate her dual chamber pulse generator so quickly after implantation that exit block occurred within 17 h. She had wound the two leads as far as their tips in a perfect formation around the pulse generator.

Myocardial sympathetic innervation in the athlete's sinus bradycardia: is there selective inferior myocardial wall denervation?

BACKGROUND

Sinus bradycardia in trained athletes is predominantly a manifestation of increased vagal tone, but it is not known whether an alteration in the cardiac sympathetic system can contribute to blunted chronotropic response. This study assessed the integrity of the sympathetic system in trained athletes with sinus bradycardia by means of the iodine-123-metaiodobenzylguanidine (123I-MIBG) procedure.

METHODS AND RESULTS

Fourteen athletes with sinus bradycardia and 8 athletes with a normal heart rate were explored by means of planar and single photon emission computed tomography MIBG studies. The heart/mediastinum ratio, regional myocardial distribution, and percent of regional myocardial MIBG uptake were evaluated. The heart/mediastinum ratio in athletes with sinus bradycardia was 1.87+/-0.10, and in athletes with a normal heart rate, the heart/mediastinum ratio was 1.86+/-0.16 (P = not significant). In athletes with sinus bradycardia, the regional distribution of MIBG showed an inferior and apical uptake defect in 8 athletes, an inferior, apical, and septal defect in 3 athletes, an inferior defect in 1 athlete, and normal distribution in 2 athletes (14%). In athletes with a normal heart rate, the regional distribution of MIBG showed an apical uptake defect in 3 athletes and normal distribution in 5 athletes (63%). The percent of regional MIBG uptake in the inferior region was significantly reduced in athletes with sinus bradycardia (44%+/-13% vs. 72%+/-11%, P<.01).

CONCLUSION

These results show severely reduced myocardial MIBG distribution in the inferior region in athletes with sinus bradycardia, suggesting selective inferior myocardial wall sympathetic denervation, which may be related to increased vagal tone.

Doppler echocardiography with extended transesophageal atrial pacing: predicting the efficacy of permanent atrial pacing in the patient with a small left ventricle and sinus node dysfunction

A 2100-g neonate underwent a two-ventricular surgical repair of a right ventricle-dominant unbalanced atrioventricular septal defect associated with the heterotaxy syndrome and sinus node dysfunction. Postoperative congestive heart failure persisted despite bradycardia management by temporary ventricular pacing. Spectral Doppler echocardiographic analysis of pulmonary venous inflow and aortic outflow patterns demonstrated significant improvement with transesophageal atrial pacing. Extended transesophageal pacing was performed for two days, resulting in dramatic clinical improvement. This is the first report of extended transesophageal atrial pacing complementing Doppler echocardiography predicting an improved outcome with permanent atrial pacing.

"Cristal tachycardias": origin of right atrial tachycardias from the crista terminalis identified by intracardiac echocardiography

OBJECTIVES

We sought to use intracardiac echocardiography (ICE) to identify the anatomic origin of focal right atrial tachycardias and to define their relation with the crista terminalis (CT).

BACKGROUND

Previous studies using ICE during mapping of atrial flutter and inappropriate sinus tachycardia have demonstrated an important relation between endocardial anatomy and electrophysiologic events. Recent studies have suggested that right atrial tachycardias may also have a characteristic anatomic distribution.

METHODS

Twenty-three consecutive patients with 27 right atrial tachycardias were included in the study. ICE was used to facilitate activation mapping in relation to endocardial structures. A 20-pole catheter was positioned along the CT under ICE guidance. ICE was also used to assist in guiding detailed mapping with the ablation catheter in the right atrium.

RESULTS

Of 27 focal right atrial tachycardias, 18 (67%, 95% confidence interval [CI] 46% to 83%) were on the CT (2 high medial, 8 high lateral, 6 mid and 2 low). ICE identified the location of the tip of the ablation catheter in immediate relation to the CT in all 18 cases. The 20-pole mapping catheter together with echocardiographic visualization of the CT provided a guide to the site of tachycardia origin along this structure. Radiofrequency ablation was successful in 26 (96%) of 27 (95% CI 81% to 100%) right atrial tachycardias.

CONCLUSIONS

This study demonstrates that approximately two thirds of focal right atrial tachycardias occurring in the absence of structural heart disease will arise along the CT. Recognition of this common distribution may potentially facilitate mapping and ablation of these tachycardias.

[Cardiomyopathy induced by tachycardia: description of a typical clinical case]

Different chronic tachyarrhythmias may cause a form of reversible myocardial dysfunction and left ventricular enlargement known as tachycardia-induced cardiomyopathy. Many authors in recent years observed that the control of heart rate or interruption of the arrhythmia were frequently associated with a significant improvement in ventricular function and a reduction of cardiac dimensions. We report a case of tachycardia-induced cardiomyopathy secondary to atrial fibrillation of unknown duration which occurred in a young patient with structurally normal heart. The regression of left ventricular enlargement and the recovery of left ventricular systolic function were complete few months after the restoration of sinus rhythm.

[Predictive factors of thromboembolic complications in mitral stenosis in sinus rythm]

The authors analyse the predisposing factors to the development of thromboembolic complications in mitral stenosis in sinus rhythm and propose preventive therapeutic measures. Eighty five consecutive patients with pure or very predominant mitral stenosis in sinus rhythm were included in this study and divided into two groups according to the presence (Group I: n = 27, age: 34.2 +/- 8.31 years) or absence (Group II: n = 58, age: 32.6 +/- 9.7 years) of thromboembolic complications. No significant difference was observed between the two groups for age, sex and functional class. Patients of group I had a more dilated left auricle (57.3 +/- 4.5 vs 48.4 +/- 4.7 mm; p < 0.001) and a smaller mitral surface area (0.8 +/- 0.15 vs 1.1 +/- 0.21 cm2, p < 0.05). The spontaneous left intra-atrial contrast phenomenon was more frequently observed in patients with thromboembolic complications (23 out of 27) than in those not presenting this complication (17 out of 58), (p < 0.001). This phenomenon was the only independent predictive factor on multivariate analysis. In conclusion, left atrial dilatation, the severity of mitral stenosis and especially the presence of spontaneous contrast are the main predictive factors of the development of thromboembolic complications in mitral stenosis in sinus rhythm. Patients presenting one or several of these factors may benefit from prophylactic anticoagulant treatment.

A new technique for the identification of respiratory sinus arrhythmia in utero

Fetal heart rate (FHR) monitoring forms the basis of routine fetal assessment, particularly short-term variability in the interbeat interval which can be difficult to interpret. Respiratory sinus arrhythmia (RSA), the change in heart rate in response to breathing, contributes to short-term variability, and the presence of RSA in utero may reflect the functional integrity of the central nervous system. This paper describes the use of Doppler ultrasound to derive the required measures of fetal heart rate and fetal breathing movements and spectral analysis to identify RSA. Cases are presented to illustrate the results obtained both in the presence and absence of RSA.

What's the price to be paid for rate response: AV sequential versus ventricular pacing?

The purpose of this study was to compare the effects of atrioventricular (AV) sequential and ventricular pacing at rest and during exercise on parameters of left ventricular performance. Twenty-five patients were studied by means of first pass radionuclide angiography. Pacing rates increased significantly (P less than 0.001) during exercise in both pacing modes, resulting in a significant increase in the cardiac index (P less than 0.001). Pulmonary transit times decreased significantly (P less than 0.001) during exercise in both pacing modes with a significantly shorter pulmonary transit time for AV sequential pacing at rest (P less than 0.01) and during exercise (P less than 0.05), indicating impaired left ventricular function in ventricular pacing. Regional left ventricular wall movement deteriorated significantly during exercise in both pacing modes (P less than 0.02), with a significantly worse performance during ventricular pacing at rest (P less than 0.05) and during exercise (P less than 0.05). Therefore, the price to be paid for rate response is a deterioration of regional wall movement. An additional loss of AV synchrony worsens the situation. It is concluded that rate modulated pacing requires preservation of AV coordination to optimize left ventricular performance.

Reliability of three computer methods in the analysis of ECG-gated radionuclide left ventriculography: interrecording, interobserver and intraobserver variability

Three different computer methods for analysis of systolic and diastolic left ventricular function (ejection fraction, peak ejection rate, time to peak ejection rate, peak filling rate, time to peak filling rate, duration of fast filling phase, and fast filling fraction) as derived from ECG-gated radionuclide cardiography were compared in 30 patients with various diseases. The patients had two gamma camera recordings of the left ventricle performed immediately following one another during radionuclide (99mTc) equilibrium (3 x 10(6) counts, 16 frames/cycle, 64 x 64 pixels). Mean ECG R-R interval of the patients remained unchanged from first to second recording. The three computer methods were: (1) end-diastolic (ED) region of interest (ROI) analysis based on manually defined ED-ROI; (2) multi (M) ROI, manually defined ROI for each frame; and (3) semiautomatic (SA) ROI, ROI for each frame defined by an SA edge detection technique. With the 16 frame points as nodes, a 160-point time-activity curve was constructed for each of the three methods by use of a spline function. A tailored multiway analysis of variance showed that the M-ROI method had the highest interindividual range of values of the function parameters and the smallest interrecording, interobserver, and intraobserver variabilities. In theory this implies a better diagnostic sensitivity and specificity for the M-ROI method as compared with the other two methods.

Mitral regurgitation and left ventricular outflow tract obstruction in hypertrophic cardiomyopathy. Analysis by Doppler echocardiography

In a group of five patients with asymmetric septal hypertrophy, the presence of the mitral regurgitation and the left ventricular outflow tract gradient were studied by the Doppler echocardiographic technique. By means of Bernoulli's formula 4V2 = P1-P2, left ventricular maximal systolic pressure was assessed using mitral regurgitant and left ventricular outflow tract jets. Aortic systolic and diastolic blood pressures were measured by cuff sphygmomanometry, and simultaneous carotid pulse tracings were recorded. All patients had outflow tract pressure gradients (peak 84 +/- 11.7 mmHg) and mild mitral regurgitation. Continuous wave Doppler study recorded peak flow velocities in the outflow tract (4.6 +/- 0.26 m/sec) and mitral regurgitant (6.5 +/- 0.51 m/sec) jets. The values for ventricular systolic pressures by both methods were similar (180 +/- 24.2 vs 186 +/- 13.5 mmHg). This study confirms the presence of mitral regurgitation and actual obstruction in hypertrophic cardiomyopathy with asymmetric septal hypertrophy.

A clinical angiographic study of the arterial blood supply to the sinus node

We studied the arterial blood supply to the sinus node area in 309 consecutive patients undergoing coronary arteriography. Seventy-nine had had a previous myocardial infarction. In two who developed temporary sinus node dysfunction, the sinus node artery arose from the distal portion of a severely stenotic left circumflex coronary artery, but in 307 of the 309 patients the sinus node arteries themselves were free of atheroma. The sinus node artery arose from the right coronary artery in 182 patients, from the proximal 3 cm in 179, near the origin of the acute marginal in one, and left of the crux in two. In 119 patients it arose from the left circumflex coronary artery, proximally in 87, and in 32 from anywhere throughout its length, running posteriorly as the posterior sinus node artery. Finally, eight patients had two sinus node arteries, one arising from the right coronary artery and one from the left circumflex.

Coronary artery spasm with sinus node dysfunction and syncope

In a 55-year-old man, attacks of spontaneous angina were associated with dizziness and syncope. Holter ECG monitoring disclosed evidence of sinus node dysfunction. Dizziness and syncope were corrected by a permanent ventricular demand pacemaker. Coronary cineangiography showed spontaneous, severe, diffuse spasm in a dominant left coronary artery and localized spasm in a nondominant right coronary artery. The patient died of pump failure shortly after cardiac catheterization. An autopsy disclosed only minimal coronary atherosclerosis. This patient's condition shows that (1) coronary spasm may cause sinus node dysfunction, dizziness, and syncope, (2) severe spasm that involves all the coronary artery branches may be fatal, and (3) severe spasm occur in minimally diseased coronary arteries confirmed by pathologic examination.