Corpus ID: 231573465

Identification of 27 cardiac abnormalities from multi-lead ECG signals: An ensembled Se-ResNet framework with Sign Loss function

  title={Identification of 27 cardiac abnormalities from multi-lead ECG signals: An ensembled Se-ResNet framework with Sign Loss function},
  author={Zhaowei Zhu and Xiang Lan and Tingting Zhao and Yangming Guo and Pipin Kojodjojo and Zhuoyang Xu and Zhuo Liu and Siqi Liu and Han Wang and Xingzhi Sun and Mengling Feng},
Objective: Cardiovascular disease is a major threat to health and one of the primary causes of death globally, where the cardiac abnormality is the most common type of cardiovascular disease. The early and accurate diagnosis of cardiac abnormalities will allow early treatment and intervention to prevent further progression of the disease. In accordance with the PhysioNet/Computing in Cardiology Challenge 2020, our objective is to develop an algorithm that automatically identifies 27 types of… Expand

Figures and Tables from this paper


Classification of 12-lead ECGs: the PhysioNet/Computing in Cardiology Challenge 2020
This Challenge provided several innovations, including a novel evaluation metric that considers different misclassification errors for different cardiac abnormalities, reflecting the clinical reality that some diagnoses have similar outcomes and varying risks. Expand
ECG signal classification for the detection of cardiac arrhythmias using a convolutional recurrent neural network.
OBJECTIVE The electrocardiogram (ECG) provides an effective, non-invasive approach for clinical diagnosis in patients with cardiac diseases such as atrial fibrillation (AF). AF is the most commonExpand
Comparing feature-based classifiers and convolutional neural networks to detect arrhythmia from short segments of ECG
This study classify short segments of ECG into four classes (AF, normal, other rhythms or noise) as part of the Physionet/Computing in Cardiology Challenge 2017 and compares a state-of-the-art feature-based classifier with a convolutional neural network approach. Expand
A convolutional neural network for ECG annotation as the basis for classification of cardiac rhythms.
This work trained a convolutional neural network with waveforms from PhysioNet databases to annotate QRS complexes, P waves, T waves, noise and interbeat ECG segments that characterize the essences of normal and irregular heart beats and identified key features for normal sinus rhythm, atrial fibrillation, alternative rhythm and noisy recordings. Expand
Arrhythmia Classification with Attention-Based Res-BiLSTM-Net
This work proposes an end-to-end deep neural network called attention-based Res-BiLSTM-Net for automatic diagnosis of cardiac arrhythmias and achieves a good result with an average F1score of 0.8757 on a multi-label arrh rhythmias classification problem in the First China ECG Intelligent Competition. Expand
Ensembling convolutional and long short-term memory networks for electrocardiogram arrhythmia detection.
An intelligent tool that assists cardiologists in identifying automatically cardiac arrhythmias and noise in electrocardiogram (ECG) recordings is constructed using a convolutional neural network and a sequence of long short-term memory units. Expand
Detection of AF and other rhythms using RR variability and ECG spectral measures
This approach seems promising in particular in detecting atrial fibrillation, and further work is needed to improve the discrimination of other rhythms and noisy signals. Expand
A 12-Lead ECG Arrhythmia Classification Method Based on 1D Densely Connected CNN
An electrocardiogram (ECG) arrhythmia classification method for short 12-lead ECG records to identify nine types, using a 1D densely connected CNN which is a relatively novel convolutional neural network (CNN) model and shows outstanding performance in the field of pattern recognition. Expand
Automatic Multi-label Classification in 12-Lead ECGs Using Neural Networks and Characteristic Points
This study proposes a novel 12-lead electrocardiogram (ECG) multi-label classification algorithm using a combination of Neural Network (NN) and the characteristic points and the end-to-end model. Expand
Classification of short single-lead electrocardiograms (ECGs) for atrial fibrillation detection using piecewise linear spline and XGBoost.
A novel methodology to classify electrocardiograms to normal, atrial fibrillation and other cardiac dysrhythmias as defined by the PhysioNet Challenge 2017, using piecewise linear splines for the feature selection and a gradient boosting algorithm for the classifier. Expand