Key result
A deep neural network trained on over 2 million 12-lead ECGs matched or outperformed medical residents and students in recognizing six types of abnormalities, achieving F1 scores above 80% and specificity over 99%.
Why the study?
The clinical utility of automatic ECG analysis has been limited by existing model accuracy, and whether deep neural networks generalize to 12-lead ECGs remained to be demonstrated.
Does a deep neural network improve the diagnostic accuracy of 12-lead ECGs for 6 common abnormalities compared to medical residents and students?
Observational (n=1,677,211)
Yes
Does a deep neural network improve the diagnostic accuracy of 12-lead ECGs for 6 common abnormalities compared to medical residents and students?
Answers are grounded in this paper and cite their sources.
A deep neural network trained on over 2 million 12-lead ECGs can accurately recognize six common rhythm and morphological abnormalities with performance matching or exceeding that of medical residents and students.
The role of automatic electrocardiogram (ECG) analysis in clinical practice is limited by the accuracy of existing models. Deep Neural Networks (DNNs) are models composed of stacked transformations that learn tasks by examples. This technology has recently achieved striking success in a variety of task and there are great expectations on how it might improve clinical practice. Here we present a DNN model trained in a dataset with more than 2 million labeled exams analyzed by the Telehealth Network of Minas Gerais and collected under the scope of the CODE (Clinical Outcomes in Digital Electrocardiology) study. The DNN outperform cardiology resident medical doctors in recognizing 6 types of abnormalities in 12-lead ECG recordings, with F1 scores above 80% and specificity over 99%. These results indicate ECG analysis based on DNNs, previously studied in a single-lead setup, generalizes well to 12-lead exams, taking the technology closer to the standard clinical practice.
No takes yet. Share an insight, caveat, or question.
A 2020 study conducted an observational in ECG abnormalities (n=1,677,211). Deep Neural Network (DNN) vs. Cardiology residents, emergency residents, and medical students was evaluated on Diagnostic accuracy (F1 score and specificity) for 6 types of ECG abnormalities. A deep neural network trained on over 2 million 12-lead ECGs matched or outperformed medical residents and students in recognizing six types of abnormalities, achieving F1 scores above 80% and specificity over 99%.
Synapse has enriched 10 related papers. Consider them for comparative context: