Key result
A dense CNN algorithm achieves an 0.80 F1 score for detecting AF and other rhythms.
Why the study?
Wearable technology enables high-quality single-channel ECG recordings, creating an opportunity for large-scale atrial fibrillation screening using short ECG segments.
Does an automatic classification algorithm using SQI and dense CNNs accurately detect atrial fibrillation from short single-lead ECG recordings?
Does an automatic classification algorithm using SQI and dense CNNs accurately detect atrial fibrillation from short single-lead ECG recordings?
A dense convolutional neural network combined with signal quality analysis achieved an F1 score of 0.80 for detecting atrial fibrillation and other rhythms from short single-lead ECGs.
Should not yet change practice; leaves open CNN utility for AF detection pending prospective clinical validation.
The development of new technology such as wearables that record high-quality single channel ECG, provides an opportunity for ECG screening in a larger population, especially for atrial fibrillation screening. The main goal of this study is to develop an automatic classification algorithm for normal sinus rhythm (NSR), atrial fibrillation (AF), other rhythms (O), and noise from a single channel short ECG segment (9-60 seconds). For this purpose, signal quality index (SQI) along with dense convolutional neural networks was used. Two convolutional neural network (CNN) models (main model that accepts 15 seconds ECG and secondary model that processes 9 seconds shorter ECG) were trained using the training data set. If the recording is determined to be of low quality by SQI, it is immediately classified as noisy. Otherwise, it is transformed to a time-frequency representation and classified with the CNN as NSR, AF, O, or noise. At the final step, a feature-based post-processing algorithm classifies the rhythm as either NSR or O in case the CNN model's discrimination between the two is indeterminate. The best result achieved at the official phase of the PhysioNet/CinC challenge on the blind test set was 0.80 (F1 for NSR, AF, and O were 0.90, 0.80, and 0.70, respectively).
No takes yet. Share an insight, caveat, or question.
Rubin et al. (2017) studied Atrial Fibrillation (n=8,528). Dense Convolutional Neural Networks and Signal Quality Analysis was evaluated on Average F1 score for classification of normal sinus rhythm, atrial fibrillation, and other rhythms. An automatic classification algorithm using dense convolutional neural networks and signal quality analysis achieved an overall F1 score of 0.80 for detecting atrial fibrillation and other rhythms on the blind test set.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: