Diffusion-Based Electrocardiography Noise Quantification via Anomaly Detection
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916855011082240 |
|---|---|
| author | Han, Tae-Seong Heo, Jae-Wook Kim, Hakseung Lee, Cheol-Hui Huh, Hyub Choi, Eue-Keun Kim, Hye Jin Kim, Dong-Joo |
| author_facet | Han, Tae-Seong Heo, Jae-Wook Kim, Hakseung Lee, Cheol-Hui Huh, Hyub Choi, Eue-Keun Kim, Hye Jin Kim, Dong-Joo |
| contents | 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 ($W_1$). Our model achieved a macro-average $W_1$ 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_11815 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Diffusion-Based Electrocardiography Noise Quantification via Anomaly Detection Han, Tae-Seong Heo, Jae-Wook Kim, Hakseung Lee, Cheol-Hui Huh, Hyub Choi, Eue-Keun Kim, Hye Jin Kim, Dong-Joo Signal Processing Artificial Intelligence Machine Learning Image and Video Processing 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 ($W_1$). Our model achieved a macro-average $W_1$ 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. |
| title | Diffusion-Based Electrocardiography Noise Quantification via Anomaly Detection |
| topic | Signal Processing Artificial Intelligence Machine Learning Image and Video Processing |
| url | https://arxiv.org/abs/2506.11815 |