scholarly journals Sensitivity and Generalization of a Neural Network for Estimating Left Atrial Fibrotic Volume Fractions from the 12-lead ECG

2021 ◽  
Vol 7 (2) ◽  
pp. 307-310
Author(s):  
Claudia Nagel ◽  
Olaf Dössel ◽  
Axel Loewe
Keyword(s):  
Neural Network ◽  
Left Atrial ◽  
Volume Fraction ◽  
Machine Learning Algorithms ◽  
Clinical Translation ◽  
P Wave ◽  
Atrial Fibrosis ◽  
Test Set ◽  
Wave Duration ◽  
P Wave Duration

Abstract Features extracted from P waves of the 12-lead electrocardiogram (ECG) have proven valuable for noninvasively estimating the left atrial fibrotic volume fraction associated with the arrhythmogenesis of atrial fibrillation. However, feature extraction in the clinical context is prone to errors and oftentimes yields unreliable results in the presence of noise. This leads to inaccurate input values provided to machine learning algorithms tailored at estimating the amount of atrial fibrosis with clinical ECGs. Another important aspect for clinical translation is the network’s generalization ability regarding new ECGs. To quantify a network’s sensitivity to inaccurately extracted P wave features, we added Gaussian noise to the features extracted from 540,000 simulated ECGs consisting of P wave duration, dispersion, terminal force in lead V1, peak-to-peak amplitudes, and additionally thoracic and atrial volumes. For assessing generalization, we evaluated the network performance for train-validation-test splits divided such that ECGs simulated with the same atria or torso geometry only belonged to either the training and validation or the test set. The root mean squared error (RMSE) of the network increased the most in case of noisy torso volumes and P wave durations. Large generalization errors with a RMSE difference between training and test set of more than 2% fibrotic volume fraction only occurred if very high or low atria and torso volumes were left out during training. Our results suggest that P wave duration and thoracic volume are features that have to be measured accurately if employed for estimating atrial fibrosis with a neural network. Furthermore, our method is capable of generalizing well to ECGs simulated with anatomical models excluded during training and thus meets an important requirement for clinical translation.

scholarly journals Evaluation of left atrial mechanics and p-wave dispersion as markers of left atrial cardiopathy in patients of embolic stroke of undetermined source (ESUS)

2020 ◽  
Vol 41 (Supplement_2) ◽  
Author(s):  
S Masood ◽  
M.M Azharuddin ◽  
S.M.K Ashraf ◽  
S Wahab
Keyword(s):  
Left Atrial ◽  
Wave Dispersion ◽  
Embolic Stroke ◽  
P Wave ◽  
Electromechanical Delay ◽  
P Wave Dispersion ◽  
Atrial Electromechanical Delay ◽  
Wave Duration ◽  
P Wave Duration ◽  
Increased Risk

Abstract Introduction Around 25% of all ischaemic strokes have no known cause despite standard investigations. Most of these cases are suspected to have an embolic source for ischaemia, leading to the moniker of Embolic Stroke of Undetermined Source (ESUS). Recent studies suggest that abnormalities of the left atrium, in the form of atrial cardiopathy, can lead to increased risk of stroke even in the absence of atrial fibrillation (AF), which may be either as precursors to AF or as independent risk factors for the development of left atrial thrombus and subsequent stroke. Purpose The aim of this study was to measure LA electromechanical dissociation, LA volumes and P-wave dispersion as markers of atrial cardiopathy in patients with ESUS to determine whether atrial cardiopathy may be in the causal pathway of ESUS. Methods 28 patients presenting with ischaemic stroke and fulfilling the criteria for ESUS were enrolled into this cross-sectional, observational study. All patients had 24-hour Holter monitoring done to rule out the presence of AF. The control group consisted of 28 age- and gender-matched apparently healthy individuals. On ECG, P-wave Dispersion (PWD) was calculated by subtracting minimum P-wave duration from maximum P-wave duration. On echocardiography, time intervals from the beginning of P-wave to beginning of A' wave from the lateral mitral annulus in tissue doppler imaging was measured as the atrial electromechanical delay. LA volumes were recorded using the Modified Biplane Simpson's method. Statistical analysis was performed using student's t-test, chi-square test, and Pearson's test. Results Baseline demographic and laboratory characteristics were similar between the two groups. Increased PWD (34.14±9.89 ms vs. 27.32±8.95 ms; p=0.01), atrial electromechanical delay (73.32±16.31 ms vs. 63.63±13.59 ms; p=0.02) and LA volumes were observed in patients with ESUS as compared to controls. A significant correlation was also found between these parameters (p<0.01). Discussion According to the results of our study, PWD, atrial electromechanical delay and LA volumes may be novel predictors for ESUS. Atrial cardiopathy is a unique mechanism of thrombo-embolism in ESUS patients and our data establishes its association with ESUS. Further studies will be needed to shed more light on its role in the causality of stroke in the ESUS population. Measurement of electromechanical delay Funding Acknowledgement Type of funding source: None

The impact of left atrial pressure on filtered P-wave duration in patients with atrial fibrillation

2016 ◽  
Vol 31 (11) ◽  
pp. 1848-1854 ◽  
Author(s):  
Hideyuki Kishima ◽  
Takanao Mine ◽  
Satoshi Takahashi ◽  
Kenki Ashida ◽  
Masaharu Ishihara ◽  
...  
Keyword(s):  
Atrial Fibrillation ◽  
Left Atrial ◽  
P Wave ◽  
Atrial Pressure ◽  
Left Atrial Pressure ◽  
Wave Duration ◽  
P Wave Duration ◽  
The Impact

Correlation of Left Atrial Size With P-Wave Duration in Interatrial Block

CHEST Journal ◽  
2005 ◽  
Vol 128 (4) ◽  
pp. 2615-2618 ◽  
Author(s):  
Vignendra Ariyarajah ◽  
Kristin Mercado ◽  
Sirin Apiyasawat ◽  
Puneet Puri ◽  
David H Spodick
Keyword(s):  
Left Atrial ◽  
P Wave ◽  
Left Atrial Size ◽  
Wave Duration ◽  
P Wave Duration ◽  
Atrial Size ◽  
Interatrial Block

scholarly journals Diagnostic performance of P wave duration in the identification of left atrial enlargement in dogs

2012 ◽  
Vol 53 (5) ◽  
pp. 267-272 ◽  
Author(s):  
P. Savarino ◽  
M. Borgarelli ◽  
A. Tarducci ◽  
S. Crosara ◽  
N. M. Bello ◽  
...  
Keyword(s):  
Left Atrial ◽  
Diagnostic Performance ◽  
P Wave ◽  
Left Atrial Enlargement ◽  
Wave Duration ◽  
P Wave Duration ◽  
Atrial Enlargement

scholarly journals P Wave Duration/P Wave Voltage Ratio Plays a Promising Role in the Prediction of Atrial Fibrillation: A New Player in the Game

2021 ◽  
Vol 2021 ◽  
pp. 1-9
Author(s):  
E. Karacop ◽  
A. Enhos ◽  
N. Bakhshaliyev ◽  
R. Ozdemir
Keyword(s):  
Left Atrial ◽  
Left Atrial Volume ◽  
P Wave ◽  
Volume Index ◽  
Atrial Volume ◽  
Left Atrial Volume Index ◽  
Wave Duration ◽  
Voltage Ratio ◽  
P Wave Duration ◽  
New Onset

Background. Atrial fibrillation (AF) is the most common sustained arrhythmia in clinical practice. Identification of patients at risk for developing AF and the opportunity for early targeted intervention might have a significant impact on morbidity and mortality. Prolonged P wave duration and decreased P wave voltage have been shown to be independent predictors of AF. The present study aimed to investigate the role of P wave duration/P wave voltage in predicting new-onset AF. Methods. We screened a total of 640 consecutive patients who admitted to cardiology outpatient clinic with a complaint of palpitation between 2012 and 2014. 24-h Holter monitoring, echocardiography, and electrocardiography (ECG) recordings were reviewed to identify new-onset AF. Patients were assigned into two groups based on presence (n = 150) and absence (n = 490) of new-onset AF. Previous ECGs with sinus rhythm were analyzed. P wave duration was measured in inferior leads, and P wave voltage was measured in lead one. P wave duration/P wave voltage was also calculated for each patient. Results. One hundred fifty subjects (23.4%) had new-onset AF among 640 patients. P wave duration (123.27 ± 12.87 vs. 119.33 ± 17.39 ms, p = 0.024 ) and P wave duration/P wave voltage (1284.70 ± 508.03 vs. 924.14 ± 462.06 ms/mV, p < 0.001 ) were higher, and P wave voltage (0.12 ± 0.04 vs. 0.13 ± 0.04 mV, p < 0.001 ) was significantly lower in the new-onset AF group compared with non-AFs. P wave duration/P wave voltage, with a cut off of 854.5 ms/mV, had 83.3% sensitivity and 62.0% specificity in a receiver operating characteristic curve (AUC 0.728, 95% CI 0.687–0.769; p < 0.001 ). Their negative and positive predictive values were 78.7% and 68.6%, respectively. In a univariate regression analysis, age, smoking, C-reactive protein, brain natriuretic peptide, left atrial diameter, left atrial volume index, P wave duration, P wave voltage, and P wave duration/P wave voltage were significantly associated with the development of new-onset AF. Moreover, smoking (OR 4.008, 95% CI 1.707–9.409; p = 0.001 ), left atrial volume index (OR 7.108, 95% CI 4.400–11.483; p < 0.001 ), and P wave duration/P wave voltage (OR 1.002, 95% CI 1.000–1.003; p = 0.044 ) were found to be significant independent predictors of new-onset AF in a multivariate analysis, after adjusting for other risk parameters. Conclusion. The P wave duration/P wave voltage ratio is a practical, easy-to-use, cheap, and reliable electrocardiographic parameter, which can play a promising role for both in predicting and elucidating a mechanism of new-onset AF.

P3838Evaluation of echocardiographic determinants of interatrial block in patients with essential hypertension

2019 ◽  
Vol 40 (Supplement_1) ◽  
Author(s):  
M Kalcik ◽  
L Bekar ◽  
O Celik ◽  
M Yetim ◽  
T Dogan ◽  
...  
Keyword(s):  
Essential Hypertension ◽  
Left Ventricular Hypertrophy ◽  
Left Ventricular Mass ◽  
Left Atrial ◽  
Ventricular Hypertrophy ◽  
Left Ventricular ◽  
P Wave ◽  
Hypertensive Patients ◽  
Wave Duration ◽  
P Wave Duration

Abstract Introduction Interatrial block (IAB), defined as a conduction delay between the right and left atrium, is manifested on the electrocardiogram as a prolonged P-wave duration. Large number of studies recently have been published regarding the prevalence of IAB and its associations with cardiovascular events. It has been previously reported that myocardial fibrosis may cause conduction delays in hypertensive patients. In this study, we aimed to investigate the echocardiographic determinants of IAB in patients with hypertension. Methods This study enrolled a total of 220 patients [male: 93 (42.3%), mean age: 61.8±7.8 years] with hypertension. Patients with atrial fibrillation were excluded. IAB was defined as P wave duration of ≥120 ms with or without presence of notching. Routinely obtained 12-lead electrocardiography recordings were examined, and patients were divided into two groups as those with and without IAB. All patients were evaluated by transthoracic echocardiography. Results Electrocardiography revealed IAB in 70 patients. The baseline demographic characteristics of the patients with and without IAB were similar in both groups. Left atrial diameter (LAD), interventricular septal thickness, posterior wall thickness, left ventricular mass, left ventricular mass index, and the prevalence of left ventricular hypertrophy (LVH) were found to be significantly increased in hypertensive patients with IAB. Increased LAD (OR=1.082; 95% CI: 1.006–1.164; p=0.034) and LVH (OR=3.051; 95% CI: 1.164–7.994; p=0.023) were found to be independent predictors of IAB. In the receiver operating characteristic curve analyses, LAD values above 38 mm predicted IAB with a sensitivity of 68%, and a specificity of 58% (AUC= 0.678; 95% CI: 0.601–0.754; p<0.001). Conclusion A significant association exists between the presence of IAB and echocardiographic parameters related to left ventricular hypertrophy and left atrial dilatation in patients with essential hypertension. The presence of IAB may be used as an electrocardiographic marker of cardiac remodelling and myocardial fibrosis in hypertensive patients.

Sign in / Sign up

Export Citation Format

Share Document