CardioPatternFormer: Pattern-Guided Attention for Interpretable ECG Classification with Transformer Architecture
Fuente:
arXiv
Salvato in:
| Autori principali: | , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866910991296495616 |
|---|---|
| author | Uğraş, Berat Kutay Gerek, Ömer Nezih Saygı, İbrahim Talha |
| author_facet | Uğraş, Berat Kutay Gerek, Ömer Nezih Saygı, İbrahim Talha |
| contents | Accurate ECG interpretation is vital, yet complex cardiac data and "black-box" AI models limit clinical utility. Inspired by Transformer architectures' success in NLP for understanding sequential data, we frame ECG as the heart's unique "language" of temporal patterns. We present CardioPatternFormer, a novel Transformer-based model for interpretable ECG classification. It employs a sophisticated attention mechanism to precisely identify and classify diverse cardiac patterns, excelling at discerning subtle anomalies and distinguishing multiple co-occurring conditions. This pattern-guided attention provides clear insights by highlighting influential signal regions, effectively allowing the "heart to talk" through transparent interpretations. CardioPatternFormer demonstrates robust performance on challenging ECGs, including complex multi-pathology cases. Its interpretability via attention maps enables clinicians to understand the model's rationale, fostering trust and aiding informed diagnostic decisions. This work offers a powerful, transparent solution for advanced ECG analysis, paving the way for more reliable and clinically actionable AI in cardiology. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_20481 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | CardioPatternFormer: Pattern-Guided Attention for Interpretable ECG Classification with Transformer Architecture Uğraş, Berat Kutay Gerek, Ömer Nezih Saygı, İbrahim Talha Signal Processing Artificial Intelligence Machine Learning Accurate ECG interpretation is vital, yet complex cardiac data and "black-box" AI models limit clinical utility. Inspired by Transformer architectures' success in NLP for understanding sequential data, we frame ECG as the heart's unique "language" of temporal patterns. We present CardioPatternFormer, a novel Transformer-based model for interpretable ECG classification. It employs a sophisticated attention mechanism to precisely identify and classify diverse cardiac patterns, excelling at discerning subtle anomalies and distinguishing multiple co-occurring conditions. This pattern-guided attention provides clear insights by highlighting influential signal regions, effectively allowing the "heart to talk" through transparent interpretations. CardioPatternFormer demonstrates robust performance on challenging ECGs, including complex multi-pathology cases. Its interpretability via attention maps enables clinicians to understand the model's rationale, fostering trust and aiding informed diagnostic decisions. This work offers a powerful, transparent solution for advanced ECG analysis, paving the way for more reliable and clinically actionable AI in cardiology. |
| title | CardioPatternFormer: Pattern-Guided Attention for Interpretable ECG Classification with Transformer Architecture |
| topic | Signal Processing Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2505.20481 |