Track topics on Twitter Track topics that are important to you
We aim to combine deep neural networks and engineered features (hand-crafted features based on medical domain knowledge) together for cardiac arrhythmia detection from short single lead ECG recordings. Approach: We propose a two-stage method named ENCASE for cardiac arrhythmia detection. The first stage is feature extraction and the second stage is classifier building. In feature extraction stage, we extract both deep features and engineered features. Deep features are obtained by modifying deep neural networks into deep feature extractor. Engineered features are extracted by summarizing existing approaches into four feature groups. Then, we propose a feature aggregation approach to combine these features together. In classifier building stage, we build multiple gradient boosting decision trees and ensemble them together to get the final detector. Results: Experiments are performed on PhysioNet/Computing in Cardiology Challenge 2017 dataset. Using F1 scores reported on hidden test set as measurement, ENCASE got 0.9117 on Normal, 0.8128 on Atrial Fibrillation, 0.7505 on Others and 0.5671 on Noise, which placed 5th in the Challenge and 8th in the follow-up ranked by average of Normal, AF and Others. When rounding to two decimal places, we placed equal 1st with three others in the Challenge and equal 2nd with seven others in the follow-up. Further experiments show that combined features perform better than individual features, and deep features show more importance scores than other features. Significance: ENCASE can benefit from both feature engineering based methods and recent deep neural networks. It is flexible and can easily assimilate the ability of new cardiac arrhythmia detection methods.
This article was published in the following journal.
Name: Physiological measurement
In this paper, a novel, automated process for constructing and initializing deep feedforward neural networks based on decision trees is presented. The proposed algorithm maps a collection of decision ...
The aim of this study was to determine whether deep convolutional neural network (DCNN)-based features can represent the difference between cardiac sarcoidosis (CS) and non-CS using polar maps.
Sensory neuroscience aims to build models that predict neural responses and perceptual behaviors, and that provide insight into the principles that give rise to them. For decades, artificial neural ne...
SchNetPack is a toolbox for the development and application of deep neural networks to the prediction of potential energy surfaces and other quantum-chemical properties of molecules and materials. It ...
We present a flexible deep convolutional neural network method for the analysis of arbitrary sized graph structures representing molecules. This method, which makes use of the Lipinski RDKit module, a...
Identifying the correct arrhythmia at the time of a clinic event including cardiac arrest is of high priority to patients, healthcare organizations, and to public health. Recent developmen...
The brain networks controlling movement are complex, involving multiple areas of the brain. Some neurological diseases, like Parkinson's disease, cause abnormalities in the brain networks....
The primary purpose of this study is to improve the quality of Magnetic Resonance Imaging in patients with heart arrhythmia. Investigators will recruit 105 patients with arrhythmia and 30 ...
We aim to experiment and implement various deep learning architectures in order to achieve human-level accuracy in Computer-aided diagnosis (CAD) systems. In particular, we are interested ...
The goal of the study is to address the unmet need of TRD patients by identifying brain networks critical for treating depression and to use next generation precision DBS with steering cap...
Agents used for the treatment or prevention of cardiac arrhythmias. They may affect the polarization-repolarization phase of the action potential, its excitability or refractoriness, or impulse conduction or membrane responsiveness within cardiac fibers. Anti-arrhythmia agents are often classed into four main groups according to their mechanism of action: sodium channel blockade, beta-adrenergic blockade, repolarization prolongation, or calcium channel blockade.
A potentially lethal cardiac arrhythmia that is characterized by uncoordinated extremely rapid firing of electrical impulses (400-600/min) in HEART VENTRICLES. Such asynchronous ventricular quivering or fibrillation prevents any effective cardiac output and results in unconsciousness (SYNCOPE). It is one of the major electrocardiographic patterns seen with CARDIAC ARREST.
Abnormally rapid heartbeat, usually with a HEART RATE above 100 beats per minute for adults. Tachycardia accompanied by disturbance in the cardiac depolarization (cardiac arrhythmia) is called tachyarrhythmia.
A computer architecture, implementable in either hardware or software, modeled after biological neural networks. Like the biological system in which the processing capability is a result of the interconnection strengths between arrays of nonlinear processing nodes, computerized neural networks, often called perceptrons or multilayer connectionist models, consist of neuron-like units. A homogeneous group of units makes up a layer. These networks are good at pattern recognition. They are adaptive, performing tasks by example, and thus are better for decision-making than are linear learning machines or cluster analysis. They do not require explicit programming.
A potent anti-arrhythmia agent, effective in a wide range of ventricular and atrial arrhythmias and tachycardias. Paradoxically, however, in myocardial infarct patients with either symptomatic or asymptomatic arrhythmia, flecainide exacerbates the arrhythmia and is not recommended for use in these patients.
Cardiology is a specialty of internal medicine. Cardiac electrophysiology : Study of the electrical properties and conduction diseases of the heart. Echocardiography : The use of ultrasound to study the mechanical function/physics of the h...