Advanced searches left 3/3

Atrial Fibrillation - Astrophysics Data System

Summarized by Plex Scholar
Last Updated: 28 June 2022

* If you want to update the article please login/register

Atrial Fibrillation Detection Using Weight-Pruned, Log-Quantised Convolutional Neural Networks

Deep neural networks are a useful tool in medical applications. However, implementation of complicated DNNs on battery-powered laptops is costly, due to high energy costs for communication. The final model had a 91. 1% model compression ratio while still maintaining high model precision of 91. 7% and less than 1% loss, which was attained by 91. 7% and less than 1%.

Source link: https://ui.adsabs.harvard.edu/abs/2022arXiv220607649C/abstract


F-Wave Extraction from Single-Lead Electrocardiogram Signals with Atrial Fibrillation by Utilizing an Optimized Resonance-Based Signal Decomposition Method

However, F-waves overlap the QRS complex and T-waves in both the time and frequency domain, making this situation a difficult one. Methods: This paper discusses an improved resonance-based signal decomposition scheme for detecting F-waves in single-lead ECG signals with atrial fibrillation. It represents the ECG signal obtained from the high-resonance and low-resonance dictionaries as a linear combination of a finite number of elements chosen from the high-resonance and low-resonance dictionaries, respectively. The tunable Q-factor wavelet transformator produces the high and low resonance dictionaries, with a high Q-factor supplying a high resonance dictionary and a low Q-factor producing a low resonance dictionary. Results: The presented method helps reduce RMSE between the extracted and the simulated F-waves relative to standard beat subtraction and principal component analysis. Conclusion: The proposed method may have the ability of F-wave extraction for handheld ECG monitoring devices, especially those with fewer leads.

Source link: https://ui.adsabs.harvard.edu/abs/2022Entrp..24..812Z/abstract


Structured variable selection: an application in identifying predictors of major bleeding among hospitalized hypertensive patients using oral anticoagulants for atrial fibrillation

Predictor identification is very important in medical research because it will enable clinicians to have a greater understanding of disease epidemiology and identify patients at risk of an outcome. Selection dependencies can help to enhance model fidelity, increase the likelihood of recovering the correct model, and increase the prediction accuracy of the resulting model. Different groups of sanctions can be punished by assigning coefficients to various groups of sanctions, which may be helpful in translating some aspects of selection dependencies into variable selection. A general framework for structured variable selection is provided, as well as a way to determine whether a penalty group adhering a set of selection dependencies. Using this previous research, we developed roadmaps to derive the grouping specification for some common selection dependencies and apply them to the challenge of creating a prediction model for major bleeding among hypertensive patients recently hospitalized for atrial fibrillation and then prescription oral anticoagulants.

Source link: https://ui.adsabs.harvard.edu/abs/2022arXiv220605337W/abstract


Prediction of Type and Recurrence of Atrial Fibrillation after Catheter Ablation via Left Atrial Electroanatomical Voltage Mapping Registration and Multilayer Perceptron Classification: A Retrospective Study

To solve these problems, we used a pre-trained 3D point cloud registration scheme and finetune it on left atrial voltage charts to find the geometric feature and align all voltage maps into the same coordinate. We convert the registered points from a multilayer perceptron classifier to predict whether patients have paroxysmal or persistent AF, and the possibility of recurrence of AF in patients in sinus rhythm in one year.

Source link: https://ui.adsabs.harvard.edu/abs/2022Senso..22.4058A/abstract


Intrinsically stretchable electrode array enabled in vivo electrophysiological mapping of atrial fibrillation at cellular resolution

Chronic atrial fibrillation at high throughput and high resolution is essential for understanding the underlying mechanism and guiding definitive intervention such as cardiac ablation, but current electrophysiological techniques are limited by poor spatial quality or electromechanical uncoupling of the beating heart.

Source link: https://ui.adsabs.harvard.edu/abs/2020PNAS..11714769L/abstract


End-to-end sensor fusion and classification of atrial fibrillation using deep neural networks and smartphone mechanocardiography

The goal of this study was to develop a new deep learning framework for detecting atrial fibrillation, one of the most common heart arrhythmias, by looking at the heart's physiological functions as shown in seismocardiography and gyrocardiography results. The notion of mechanocardiography, developed jointly by SCG and GCG, is a test used to record precordial vibrations in smartphones' built-in inertial sensors. A version of the classification of sinus rhythm, AFib, and Noise categories from tri-axial SCG and GCG data derived from smartphones is shown in this paper. In addition, we use bidirectional long-term memory layers on top of fully connected layers to obtain both spatial and spatiotemporal information of the multidimensional SCG and GCG signals, as well as spatial and spatiotemporal characteristics. Our system not only can efficiently fuse SCG and GCG signals, but also can detect heart rhythms and abnormalities in the MCG signals with utmost precision.

Source link: https://ui.adsabs.harvard.edu/abs/2022PhyM...43e5004M/abstract

* Please keep in mind that all text is summarized by machine, we do not bear any responsibility, and you should always check original source before taking any actions

* Please keep in mind that all text is summarized by machine, we do not bear any responsibility, and you should always check original source before taking any actions