Detection of Atrial Fibrillation in Holter ECG Recordings by ECHOView Images: A Deep Transfer Learning Study

Vessela Krasteva, Todor Stoyanov, Stefan Naydenov, Ramun Schmid, Irena JekovaDiagnostics 2025 · DOI

TL;DR

This study leverages deep transfer learning with ImageNet-trained neural networks to accurately detect Atrial Fibrillation (AF) from novel ECHOView images derived from Holter ECG recordings. The approach achieves high accuracy, even with short ECG segments and noisy data, and provides interpretable visualizations to understand the model's decision-making process.

What Problem Does This Paper Address?

Atrial Fibrillation (AF) is a common heart arrhythmia that significantly increases the risk of stroke and other serious complications. Early and accurate detection is crucial for timely treatment. However, identifying AF, especially transient or asymptomatic episodes, in long-term (24-72 hour) Holter ECG recordings presents a significant challenge. Symptoms can be infrequent, and the sheer volume of data makes manual review time-consuming and prone to human error. Existing automated methods can struggle with noise, artifacts, and the subtle nature of short AF events. This paper aims to improve AF detection by exploring the use of deep learning on a novel visual representation of ECG data called ECHOView images.

Methods

The core of this research lies in deep transfer learning. The authors took 18 pre-trained deep neural networks (DNNs) that were originally trained on the massive ImageNet dataset (for image classification). These networks were adapted for AF detection in two main ways:

  1. Retraining: Only the final classification layers of the DNNs were modified and retrained. The earlier layers, which learned general image features, were kept "frozen." This approach is computationally efficient and requires less data.
  2. Fine-tuning: After retraining, the models were further refined by allowing some of the earlier layers to be trained with a very low learning rate. This allows the model to adapt more specifically to the ECG data while still leveraging the pre-trained knowledge.

The input to these DNNs were ECHOView images. These are 2D representations of ECG heartbeats, where amplitude is mapped to color. This visual format is designed to make arrhythmias more easily detectable. The authors generated these images from 30-second ECG segments and then rescaled them to fit the input requirements of the ImageNet DNNs.

To understand why the models made certain predictions, the researchers employed GradCAM (Gradient-weighted Class Activation Mapping). This technique generates "heatmaps" that highlight the regions within the ECHOView images that were most influential in the DNN's decision to classify a segment as AF or non-AF.

Datasets

The primary dataset used in this study is the IRIDIA-AF database. This is a publicly available Holter ECG monitoring database containing recordings from 152 patients with paroxysmal AF. The database includes:

  • 2-lead ECG signals sampled at 200 Hz.
  • Annotations for QRS positions, and onset/offset of AF episodes, reviewed by expert cardiologists.
  • Approximately 6690 hours of ECG data, with 1609 hours of recorded AF rhythms.

The authors created training, validation, and test sets from this database, ensuring a balanced split between AF and non-AF cases. The test set was kept entirely independent from the training and validation data.

Key Insights and Ideas

  • ECHOView as a powerful ECG representation: The study demonstrates that transforming ECG signals into ECHOView images can effectively capture crucial information for AF detection, making it suitable for image-based deep learning models.
  • Transfer learning is highly effective for AF detection: Leveraging pre-trained ImageNet models significantly reduces the need for massive, specialized ECG datasets and computational resources, while still achieving state-of-the-art performance.
  • Fine-tuning offers a performance boost: While retraining alone yields good results, fine-tuning the models further improves accuracy, especially for detecting AF.
  • Interpretability is key: GradCAM heatmaps provide valuable insights, showing that the DNNs focus on ECG regions that align with what cardiologists look for, enhancing trust and understanding of the automated diagnosis.
  • Robustness across various conditions: The models show remarkable resilience to variations in patient data, ECG leads, and even noise and artifacts common in real-world Holter recordings.

Key Results

  • High Accuracy: Fine-tuned models achieved accuracies of 97.2-97.6% on the test dataset. Retrained models reached 96.3%.
  • Minimal Inter-patient and Inter-lead Drops: The best models showed very small drops in accuracy when tested on different patients (1-1.6%) or different ECG leads (0.3-1.2%), indicating strong generalization.
  • Short-term Detection: Models could process shorter ECG episodes (20 seconds) with only a minor accuracy drop (up to 0.6%), and even 10-second episodes with a tolerable drop (4-15%).
  • Noise Robustness: Performance remained high even on full-length Holter recordings containing noise and rhythm variations.
  • Top Performing Models: EfficientNetV2B1 (retrained and fine-tuned) and fine-tuned DenseNet-121, -169, -201 were identified as the top performers, balancing accuracy with computational efficiency.

Limitations and Future Work

  • Database Limitations: The IRIDIA-AF database primarily contains paroxysmal AF. The study notes that the presence of other AF types (persistent, permanent) or mixed annotations (like atrial flutter) in training data could influence model performance and interpretation.
  • Disputable Cases: The study highlights cases where the DNN's decisions, while highly confident, were debatable upon expert cardiologists' review, particularly in scenarios with rapid heart rates, artifacts, or mixed rhythms. This suggests that while the models are robust, subtle distinctions and complex arrhythmias can still pose challenges.
  • Future Work: The authors suggest exploring hybrid architectures, training on larger and more diverse datasets, and testing the approach in the context of multi-label classification for a broader range of arrhythmias. They also emphasize the need for further validation and regulatory considerations for real-world deployment.

Summary updated

Abstract

<b>Background/Objectives</b>: The timely and accurate detection of atrial fibrillation (AF) is critical from a clinical perspective. Detecting short or transient AF events is challenging in 24-72 h Holter ECG recordings, especially when symptoms are infrequent. This study aims to explore the potential of deep transfer learning with ImageNet deep neural networks (DNNs) to improve the interpretation of short-term ECHOView images for the presence of AF. <b>Methods</b>: Thirty-second ECHOView images, composed of stacked heartbeat amplitudes, were rescaled to fit the input of 18 pretrained ImageNet DNNs with the top layers modified for binary classification (AF, non-AF). Transfer learning provided both retrained DNNs by training only the top layers (513-2048 trainable parameters) and fine-tuned DNNs by slowly training retrained DNNs (0.38-23.48 M parameters). <b>Results</b>: Transfer learning used 13,536 training and 6624 validation samples from the two leads in the IRIDIA-AF Holter ECG database, evenly split between AF and non-AF cases. The top-ranked DNNs evaluated on 11,400 test samples from independent records are the retrained EfficientNetV2B1 (96.3% accuracy with minimal inter-patient (1%) and inter-lead (0.3%) drops), and fine-tuned EfficientNetV2B1 and DenseNet-121, -169, -201 (97.2-97.6% accuracy with inter-patient (1.4-1.6%) and inter-lead (0.5-1.2%) drops). These models can process shorter ECG episodes with a tolerable accuracy drop of up to 0.6% for 20 s and 4-15% for 10 s. Case studies present the GradCAM heatmaps of retrained EfficientNetV2B1 overlaid on raw ECG and ECHOView images to illustrate model interpretability. <b>Conclusions</b>: In an extended deep transfer learning study, we validate that ImageNet DNNs applied to short-term ECHOView images through retraining and fine-tuning can significantly enhance automated AF diagnoses. GradCAM heatmaps provide meaningful model interpretability, highlighting ECG regions of interest aligned with cardiologist focus.
BibTeX
@inproceedings{krasteva2025detection,
  author = {Vessela Krasteva and Todor Stoyanov and Stefan Naydenov and Ramun Schmid and Irena Jekova},
  title = {{Detection of Atrial Fibrillation in Holter ECG Recordings by ECHOView Images: A Deep Transfer Learning Study}},
  booktitle = {Diagnostics},
  year = {2025},
  doi = {10.3390/diagnostics15070865},
}
Figure