aiqbench analysis
This study demonstrates that domain-knowledge-driven synthetic ECG data can significantly enhance deep neural network classification performance, particularly when real-world datasets are scarce. By pre-training models with Gaussian-composed synthetic heartbeats, the approach yields substantial accuracy gains for specific arrhythmias like atrial flutter without requiring extensive privacy-sensitive data collection. The findings offer a practical pathway for addressing data scarcity in medical AI through physics-informed simulation rather than relying solely on large-scale real-world datasets.
Boosting ECG Classification Performance by Pre-training with Synthesized Data
arXiv:2606.10802v1 Announce Type: new Abstract: Deep Neural Networks (DNNs) typically require extensive datasets for effective training. In the medical domain, acquiring large-scale data is often challenging due to privacy concerns and the rarity of certain diseases. To address this data scarcity, we investigate the efficacy of training DNN models using synthetic data, generated based on domain-specific medical knowledge. Specifically, we develop a knowledge-driven Gaussian-composition synthesis algorithm for single-lead II ECGs, in which each heartbeat is represented by Gaussian-shaped P, Q, R, S, and T wave components. Using this simulator, we generate synthetic data for four abnormal electrocardiogram (ECG) classes: atrial fibrillation (AF), atrial flutter (AFLT), premature ventricular complex (PVC), and Wolff-Parkinson-White Syndrome (WPW). We evaluate the utility of this synthetic data by conducting abnormal ECG classification using ten different DNN architectures. Our results demonstrate that synthetic-to-real training improves classification performance for three of the four target abnormalities, with the largest architecture-averaged gain of $33.2\%$ observed for AFLT. Further analysis reveals that the performance enhancement from synthetic data is more pronounced with smaller real-world datasets. These findings suggest that domain-knowledge-based synthetic ECGs can serve as a useful pre-training resource, particularly in scenarios where real-world data are limited or difficult to obtain.
Abstract:Deep Neural Networks (DNNs) typically require extensive datasets for effective training. In the medical domain, acquiring large-scale data is often challenging due to privacy concerns and the rarity of certain diseases. To address this data scarcity, we investigate the efficacy of training DNN models using synthetic data, generated based on domain-specific medical knowledge. Specifically, we develop a knowledge-driven Gaussian-composition synthesis algorithm for single-lead II ECGs, in which each heartbeat is represented by Gaussian-shaped P, Q, R, S, and T wave components. Using this simulator, we generate synthetic data for four abnormal electrocardiogram (ECG) classes: atrial fibrillation (AF), atrial flutter (AFLT), premature ventricular complex (PVC), and Wolff-Parkinson-White Syndrome (WPW). We evaluate the utility of this synthetic data by conducting abnormal ECG classification using ten different DNN architectures. Our results demonstrate that synthetic-to-real training improves classification performance for three of the four target abnormalities, with the largest architecture-averaged gain of $33.2\%$ observed for AFLT. Further analysis reveals that the performance enhancement from synthetic data is more pronounced with smaller real-world datasets. These findings suggest that domain-knowledge-based synthetic ECGs can serve as a useful pre-training resource, particularly in scenarios where real-world data are limited or difficult to obtain.
Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
arXiv:2606.10802 [cs.LG]
(or arXiv:2606.10802v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2606.10802 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 292 words
Related coverage
CLARK: Closed-loop Learning for Adaptive Reasoning over Knowledge Graphs
arXiv · Jul 23, 2026
Analyzing Image Encoder Choices and Graph Homophily in GCN Frameworks for Breast Ultrasound Classification
arXiv · Jul 30, 2026
Fourier Analysis of Parametrized Interactive Quantum Classifiers
arXiv · Sep 17, 2026
IECG: Weakly-Supervised Interpretability-Enhancing Foundation Model for Electrocardiogram
arXiv · Oct 21, 2026