{"ID":2869817,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.14302","arxiv_id":"2509.14302","title":"D4PM: A Dual-branch Driven Denoising Diffusion Probabilistic Model with Joint Posterior Diffusion Sampling for EEG Artifacts Removal","abstract":"Artifact removal is critical for accurate analysis and interpretation of Electroencephalogram (EEG) signals. Traditional methods perform poorly with strong artifact-EEG correlations or single-channel data. Recent advances in diffusion-based generative models have demonstrated strong potential for EEG denoising, notably improving fine-grained noise suppression and reducing over-smoothing. However, existing methods face two main limitations: lack of temporal modeling limits interpretability and the use of single-artifact training paradigms ignore inter-artifact differences. To address these issues, we propose D4PM, a dual-branch driven denoising diffusion probabilistic model that unifies multi-type artifact removal. We introduce a dual-branch conditional diffusion architecture to implicitly model the data distribution of clean EEG and artifacts. A joint posterior sampling strategy is further designed to collaboratively integrate complementary priors for high-fidelity EEG reconstruction. Extensive experiments on two public datasets show that D4PM delivers superior denoising. It achieves new state-of-the-art performance in EOG artifact removal, outperforming all publicly available baselines. The code is available at https://github.com/flysnow1024/D4PM.","short_abstract":"Artifact removal is critical for accurate analysis and interpretation of Electroencephalogram (EEG) signals. Traditional methods perform poorly with strong artifact-EEG correlations or single-channel data. Recent advances in diffusion-based generative models have demonstrated strong potential for EEG denoising, notably...","url_abs":"https://arxiv.org/abs/2509.14302","url_pdf":"https://arxiv.org/pdf/2509.14302v1","authors":"[\"Feixue Shao\",\"Xueyu Liu\",\"Yongfei Wu\",\"Jianbo Lu\",\"Guiying Yan\",\"Weihua Yang\"]","published":"2025-09-17T10:58:51Z","proceeding":"eess.IV","tasks":"[\"eess.IV\"]","methods":"[\"Diffusion Model\"]","has_code":false,"code_links":[{"ID":609723,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2869817,"paper_url":"https://arxiv.org/abs/2509.14302","paper_title":"D4PM: A Dual-branch Driven Denoising Diffusion Probabilistic Model with Joint Posterior Diffusion Sampling for EEG Artifacts Removal","repo_url":"https://github.com/flysnow1024/D4PM","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}