No abstract

Aims: Electrocardiographic waveforms (ECG) are recognized as the most reliable method to detect abnormal heart rhythms such as atrial fibrillation. This task is challenging when the signals are distorted by noise. This paper presents an automatic classification algorithm to classify short lead ECGs in terms of abnormality of heart rhythm (AF or alternative rhythms) and quality (noisy recordings). Methods: To meet this end, at first baseline wander removal and Butterworth filter for each signal are applied as a preprocessing stage. Due to the existence of noise in recordings, high quality beats are selected for any further analysis using cycle quality assessment. Then, three sets of features defined as correlation coefficient, fractal dimension and variance of R peaks are extracted to predict noisy recordings. Two separate approaches are employed to classify other three classes. The first approach is the feature based methodology and the second one is the applying deep neural networks. In the first approach, features from different domains are extracted .The method for AF detection utilizes and characterizes variability in RR-intervals which are extracted by applying classic Pan-Tompkins algorithm. To improve the accuracy of the AFdetection, atrial activity is analyzed by understanding whether the P-wave is present in signal. This is done by investigating the morphology of P-waves. Heart rate abnormality and the existence of premature beats in a signal are regarded as two characteristics to distinguish non-AF rhythms. The whole sets of features are fed into a neural network classifier. Another approach uses the segments with 600 samples as the input of a 1 dimensional convolutional neural network. The output obtained from both approaches are combined using a decision table and finally the recordings are classified into three classes. Results: The proposed method is evaluated using scoring function from 2017 PhysioNet/CinC Challenge and achieved an overall score of 80% and 71% on the training dataset and hidden test dataset, respectively.

No abstract

Objective: Detection of Atrial fibrillation (AF); as a very common cardiac arrhythmia; is a challenging issue because it is often asymptotic. Most of previous studies were based on feature extraction or training CNNs from scratch. Difficulties in finding appropriate features, requiring an expertise for feature extraction and tuning parameters, and needs for a large amount of labeled data are the most drawbacks of previous studies. The transfer learning is a solution for these problems especially in the medical analysis where available data is limited. The main goal of this study is to investigate the ability of transfer learning method for classification ECG signals into normal, AF, other rhythms and noisy classes. Approach: In our analysis, ECG signals have been transformed to 6 s segments images and fed to three well-known pre-trained models (AlexNet, VGG-16, and ResNet-152). Features have been extracted from different layers of the models and used as the inputs of a classifier for AF detection. Then we have compared the best resulting model to proposed models trained from scratch. The proposed method also investigated the impact of models depth. The algorithms have been trained and validated using the public dataset of 2017 Physionet challenge. Main results: The method achieved average accuracy 87.9, and F-score 86.4 on the validation dataset (publicly available dataset). The second level of extracted features has the highest accuracy for all the pre-trained models in our study. Pre-trained AlexNet outperformed full training CNNs evaluated in this study. Significance: It is shown that the transfer learning method yields good accuracy recognition performance even when source and target datasets are completely different. The results show that the transfer learning is a reliable, accurate and low computational method to classify AF from short ECG signals while requires neither feature selection nor heavy processing steps.

No abstract