Diffusion-Based Electrocardiography Noise Quantification via Anomaly Detection

2025/06/13 by Han, Tae-Seong, Hakseung Kim, Heo, Jae-Wook +11 · 1 citation
Computer Science · Medicine · #Anomaly Detection Techniques and Applications #Artificial Intelligence (cs.AI) #ECG Monitoring and Analysis #FOS: Computer and information sciences #FOS: Electrical engineering #Image and Signal Denoising Methods #Image and Video Processing (eess.IV) #Machine Learning (cs.LG) #Signal Processing (eess.SP) #electronic engineering #information engineering

paper · pdf · doi:10.48550/arxiv.2506.11815

openalex publication_date 2025/06/13 · openalex created_date 2025/10/11 · openalex updated_date 2026/07/28

Abstract

Electrocardiography (ECG) signals are frequently degraded by noise, limiting their clinical reliability in both conventional and wearable settings. Existing methods for addressing ECG noise, relying on artifact classification or denoising, are constrained by annotation inconsistencies and poor generalizability. Here, we address these limitations by reframing ECG noise quantification as an anomaly detection task. We propose a diffusion-based framework trained to model the normative distribution of clean ECG signals, identifying deviations as noise without requiring explicit artifact labels. To robustly evaluate performance and mitigate label inconsistencies, we introduce a distribution-based metric using the Wasserstein-1 distance (W₁). Our model achieved a macro-average W₁ score of 1.308, outperforming the next-best method by over 48%. External validation confirmed strong generalizability, facilitating the exclusion of noisy segments to improve diagnostic accuracy and support timely clinical intervention. This approach enhances real-time ECG monitoring and broadens ECG applicability in digital health technologies.

Citations

• A Review on Generative AI Models for Synthetic Medical Text, Time Series, and Longitudinal Data
• Learning under Label Noise through Few-Shot Human-in-the-Loop Refinement
• On Diffusion Modeling for Anomaly Detection
• SimpleNet: A Simple Network for Image Anomaly Detection and Localization
• Denoising diffusion models for out-of-distribution detection
• Automatic Detection of Noisy Electrocardiogram Signals without Explicit Noise Labels
• Diffusion Models for Medical Anomaly Detection
• Training language models to follow instructions with human feedback
• Do autoencoders need a bottleneck for anomaly detection?
• High-Resolution Image Synthesis with Latent Diffusion Models
• High-Resolution Image Synthesis with Latent Diffusion Models
• Diffusion Models Beat GANs on Image Synthesis
• PaDiM: a Patch Distribution Modeling Framework for Anomaly Detection and\n Localization
• Denoising Diffusion Implicit Models
• Same Same But DifferNet: Semi-Supervised Defect Detection with\n Normalizing Flows
• NeuroKit2: A Python toolbox for neurophysiological signal processing
• Denoising Diffusion Probabilistic Models
• Image quality assessment: from error visibility to structural similarity

Cited by

• Simulator and Experience Enhanced Diffusion Model for Comprehensive ECG Generation

Related

• Investigating the Generalizability of ECG Noise Detection Across Diverse Data Sources and Noise Types
• TFCDiff: Robust ECG Denoising via Time-Frequency Complementary Diffusion
• Enhancing ECG Classification Robustness with Lightweight Unsupervised Anomaly Detection Filters
• Representing and Denoising Wearable ECG Recordings
• Automatic Detection of Noisy Electrocardiogram Signals without Explicit Noise Labels