Key result
Varying sampling rates between training and testing reduces DL-based ECG classifier accuracy by ~20%.
Why the study?
Varying ECG sampling rates can impact deep learning models and hinder deployment of generalized classifiers across diverse instruments, but how differing sampling rates affect performance was not well understood.
Does varying sampling rates between training and testing data affect the accuracy of deep learning-based ECG classifiers?
Does varying sampling rates between training and testing data affect the accuracy of deep learning-based ECG classifiers?
Differences in sampling rates between training and testing data can significantly degrade the performance of deep learning ECG classifiers, highlighting the need for strategies like resampling or transfer learning.
Sampling rate mismatches warrant caution when deploying ECG deep learning models; leaves open optimal standardization strategies for future classifiers.
With recent advances in Deep Learning (DL) models, the healthcare domain has seen an increased adoption of neural networks for clinical diagnosis, monitoring, and prediction. Deep Learning models have been developed for various tasks using 1D (one-dimensional) time-series signals. Time-series healthcare data, typically collected through sensors, have specific structures and characteristics such as frequency and amplitude. The nature of these features, including varying sampling rates that depend on the instruments used for sensing, poses challenges in handling them. Electrocardiograms (ECG), a class of 1D time-series signals representing the electrical activity of the heart, have been used to develop heart condition classification decision support systems. The sampling rate of these signals, influenced by different ECG instruments as well as their calibrations, can greatly impact the learning functions of deep learning models and subsequently, their decision outcomes. This hinders the development and deployment of generalized, DL-based ECG classifiers that can work with data from a variety of ECG instruments, particularly when the sampling rate of the training data remains unknown to users. Moreover, DL models are not designed to recognize the sampling rate of the testing data on which they are being deployed, further complicating their effective application across diverse clinical settings. In this study, we investigated the effect of different sampling rates of time-series ECG signals on DL-based ECG classifiers. To the best of our knowledge, this is the first work to understand how varying sampling rates affect the performance of DL-based models for classifying 1D time-series ECG signals. Through our comprehensive experiments, we showed that accuracy can drop by as much as 20% when the training and testing sampling rates are different. We provide visual explanations to understand the differences in learned model features through activation maps when the sampling rates for training and testing data are different. We also investigated potential strategies to address the challenges posed by different sampling rates: (i) transfer learning, (ii) resampling, and (iii) training a DL model using ECG data at different sampling rates.
No takes yet. Share an insight, caveat, or question.
Dakshit et al. (2025) studied Heart condition classification (ECG). Varying sampling rates of time-series ECG signals vs. Matching training and testing sampling rates was evaluated on Accuracy of DL-based ECG classifiers (accuracy drop by as much as 20%). Varying sampling rates between training and testing data can reduce the accuracy of deep learning-based ECG classifiers by up to 20%.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: