{"ID":2875847,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.01330","arxiv_id":"2509.01330","title":"Prior-Guided Residual Diffusion: Calibrated and Efficient Medical Image Segmentation","abstract":"Ambiguity in medical image segmentation calls for models that capture full conditional distributions rather than a single point estimate. We present Prior-Guided Residual Diffusion (PGRD), a diffusion-based framework that learns voxel-wise distributions while maintaining strong calibration and practical sampling efficiency. PGRD embeds discrete labels as one-hot targets in a continuous space to align segmentation with diffusion modeling. A coarse prior predictor provides step-wise guidance; the diffusion network then learns the residual to the prior, accelerating convergence and improving calibration. A deep diffusion supervision scheme further stabilizes training by supervising intermediate time steps. Evaluated on representative MRI and CT datasets, PGRD achieves higher Dice scores and lower NLL/ECE values than Bayesian, ensemble, Probabilistic U-Net, and vanilla diffusion baselines, while requiring fewer sampling steps to reach strong performance.","short_abstract":"Ambiguity in medical image segmentation calls for models that capture full conditional distributions rather than a single point estimate. We present Prior-Guided Residual Diffusion (PGRD), a diffusion-based framework that learns voxel-wise distributions while maintaining strong calibration and practical sampling effici...","url_abs":"https://arxiv.org/abs/2509.01330","url_pdf":"https://arxiv.org/pdf/2509.01330v2","authors":"[\"Fuyou Mao\",\"Beining Wu\",\"Yanfeng Jiang\",\"Han Xue\",\"Yan Tang\",\"Hao Zhang\"]","published":"2025-09-01T10:13:15Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[\"Diffusion Model\"]","has_code":false}