---
title: "DiffECG: Diffusion Model-Powered Label-Efficient and Personalized Arrhythmia Diagnosis"
authors:
  - "Tianren Zhou"
  - "Zhenge Jia"
  - "Dongxiao Yu"
  - "Zhaoyan Shen"
doi: "10.24963/ijcai.2025/890"
url: "https://ijcai.org/proceedings/2025/890"
pdf: "https://ijcai.org/proceedings/2025/0890.pdf"
venue: "Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence"
year: 2025
pages: "8003-8011"
track: "Main Track"
keywords:
  - "Machine Learning: ML: Applications"
  - "Machine Learning: ML: Classification"
  - "AI Ethics, Trust, Fairness: ETF: Fairness and diversity"
  - "Multidisciplinary Topics and Applications: MTA: Health and medicine"
---

# DiffECG: Diffusion Model-Powered Label-Efficient and Personalized Arrhythmia Diagnosis

Tianren Zhou, Zhenge Jia, Dongxiao Yu, Zhaoyan Shen

*Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence*, pages 8003-8011. DOI: [10.24963/ijcai.2025/890](https://doi.org/10.24963/ijcai.2025/890)

## Abstract

Arrhythmia diagnosis using electrocardiogram (ECG) is critical for preventing cardiovascular risks. However, existing deep learning-based methods struggle with label scarcity and contrastive learning-based methods suffer from false-negative samples, which lead to poor model generalization. Besides, due to inter-subject variability, pre-trained models cannot achieve evenly performance across individuals. Conducting model fine-tuning for each individual is computationally expensive and does not guarantee improvement. We propose DiffECG, a diffusion-based self-supervised learning framework for label-efficient and personalized arrhythmia detection. Our method utilizes a diffusion model to extract robust ECG representations, coupled with a novel feature extractor and a multi-modal feature fusion strategy to obtain a well-generalized model. Moreover, we propose an efficient model personalization mechanism based on zeroth-order optimization. It personalizes the model by tuning the noise-adding step t in the diffusion process, significantly reducing computational costs compared to model fine-tuning. Experimental results show that our proposed method outperforms the SOTA method by 37.9% and 23.9% in generalization and personalization performance, respectively. The source code is available at: https://github.com/Auguuust/DiffEC