Key result
Retrained hybrid deep learning model achieves a ~0.96 F1-score for automated AF detection.
Why the study?
Automated detection of AF from ECG traces remains challenging and is crucial for telemonitoring post-stroke patients, prompting evaluation of the generalizability of a deep learning automated ECG classification algorithm.
Does a deep learning-based automated ECG classification algorithm accurately detect atrial fibrillation from short-term single-lead ECG traces?
Observational (n=102)
Yes
Does a deep learning-based automated ECG classification algorithm accurately detect atrial fibrillation from short-term single-lead ECG traces?
A deep learning-based algorithm can accurately detect atrial fibrillation from short-term single-lead ECG traces, with performance significantly improving after retraining on the target dataset.
Novel HDL model enables AF classification from ECG; leaves open external validation for post-stroke telemonitoring.
Automated detection of atrial fibrillation (AF) from electrocardiogram (ECG) traces remains a challenging task and is crucial for telemonitoring of patients after stroke. This study aimed to quantify the generalizability of a deep learning (DL)-based automated ECG classification algorithm. We first developed a novel hybrid DL (HDL) model using the PhysioNet/CinC Challenge 2017 (CinC2017) dataset (publicly available) that can classify the ECG recordings as one of four classes: normal sinus rhythm (NSR), AF, other rhythms (OR), and too noisy (TN) recordings. The (pre)trained HDL was then used to classify 636 ECG samples collected by our research team using a handheld ECG device, CONTEC PM10 Portable ECG Monitor, from 102 (age: 68 ± 15 years, 74 male) outpatients of the Eastern Heart Clinic and inpatients in the Cardiology ward of Prince of Wales Hospital, Sydney, Australia. The proposed HDL model achieved average test F <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</inf> -score of 0.892 for NSR, AF, and OR, relative to the reference values, on the CinC2017 dataset. The HDL model also achieved an average F <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</inf> -score of 0.722 (AF: 0.905, NSR: 0.791, OR: 0.471 and TN: 0.342) on the dataset created by our research team. After retraining the HDL model on this dataset using a 5-fold cross validation method, the average F <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</inf> -score increased to 0.961. We finally conclude that the generalizability of the HDL-based algorithm developed for AF detection from short-term single-lead ECG traces is acceptable. However, the accuracy of the pre-trained DL model was significantly improved by retraining the model parameters on the new dataset of ECG traces.
No takes yet. Share an insight, caveat, or question.
Argha et al. (2023) conducted an observational in Atrial fibrillation (n=102). Hybrid deep learning (HDL) model vs. Reference values was evaluated on Average F1-score for ECG classification. A hybrid deep learning model for automated atrial fibrillation detection achieved an average F1-score of 0.722 on a new clinical dataset, which improved to 0.961 after retraining.
Synapse has enriched 3 closely related papers on similar clinical questions. Consider them for comparative context: