Key result
A DenseNet121 deep learning model using reverse polar transformed spectrograms of Pan-Tompkins preprocessed ECG signals achieved a macro F1-score of 0.8238 and an accuracy of 0.9076 for classifying atrial fibrillation.
Why the study?
Wearable ECG devices increasingly monitor heart rhythms, and combining signal visualization with AI abnormality detection could help users independently assess physiological signals.
Does a deep learning model using reverse polar transformed ECG spectrograms accurately predict atrial fibrillation in single-lead ECG recordings?
Does a deep learning model using reverse polar transformed ECG spectrograms accurately predict atrial fibrillation in single-lead ECG recordings?
Reverse polar transformed ECG spectrograms combined with deep learning models provide a compact and effective method for predicting atrial fibrillation, achieving comparable performance to existing methods.
Answers are grounded in this paper and cite their sources.
Polar ECG spectrograms may support wearable arrhythmia screening; leaves open prospective validation before clinical adoption.
Portable and wearable electrocardiogram (ECG) devices are increasingly utilized in healthcare for monitoring heart rhythms and detecting cardiac arrhythmias or other heart conditions. The integration of ECG signal visualization with AI-based abnormality detection empowers users to independently and confidently assess their physiological signals. In this study, we investigated a novel method for visualizing ECG signals using polar transformations of short-time Fourier transform (STFT) spectrograms and evaluated the performance of deep convolutional neural networks (CNNs) in predicting atrial fibrillation from these polar transformed spectrograms. The ECG data, which are available from the PhysioNet/CinC Challenge 2017, were categorized into four classes: normal sinus rhythm, atrial fibrillation, other rhythms, and noise. Preprocessing steps included ECG signal filtering, STFT-based spectrogram generation, and reverse polar transformation to generate final polar spectrogram images. These images were used as inputs for deep CNN models, where three pre-trained deep CNNs were used for comparisons. The results demonstrated that deep learning-based predictions using polar transformed spectrograms were comparable to existing methods. Furthermore, the polar transformed images offer a compact and intuitive representation of rhythm characteristics in ECG recordings, highlighting their potential for wearable applications.
No takes yet. Share an insight, caveat, or question.
Kwon et al. (2025) studied Atrial fibrillation (n=5,977). DenseNet121 deep learning model with reverse polar transformed spectrograms of Pan-Tompkins preprocessed ECG signals vs. DenseNet121 model without Pan-Tompkins preprocessing was evaluated on Macro F1-score for arrhythmia classification. A DenseNet121 deep learning model using reverse polar transformed spectrograms of Pan-Tompkins preprocessed ECG signals achieved a macro F1-score of 0.8238 and an accuracy of 0.9076 for classifying atrial fibrillation.
Synapse has enriched 10 related papers. Consider them for comparative context: