Search papers, labs, and topics across Lattice.
This paper introduces Neighbor-Guided Patch Sampling (NGPS), a novel framework for self-supervised denoising in volumetric medical imaging that effectively addresses inter-slice misalignment without the need for explicit registration. By decoupling structure matching from signal retrieval, NGPS utilizes a guide image to identify structurally similar patches while directly sourcing supervision from raw noisy neighbors, thus minimizing the impact of misleading targets. Experimental results demonstrate that NGPS significantly enhances fidelity and structure-sensitive metrics in both CT and synthetic-Rician MRI datasets, outperforming prior masking approaches.
NGPS achieves superior denoising by leveraging local structural similarity without the pitfalls of traditional registration methods, leading to clearer anatomical boundaries in medical imaging.
Neighboring-slice self-supervised denoising is attractive for volumetric medical imaging, yet inter-slice misalignment breaks anatomical correspondence and often yields ghosting and blurred margins when adjacent slices are used naively as targets. We propose Neighbor-Guided Patch Sampling (NGPS), a lightweight framework that constructs neighboring supervision under local inter-slice misalignment without explicit registration. To avoid learning from misleading targets, prior methods commonly mask discrepant regions, but this stabilizes training at the cost of leaving a non-trivial portion of neighboring evidence unexploited, particularly around high-frequency anatomical boundaries. NGPS addresses this by decoupling structure matching from signal retrieval: for each masked location, it searches a local neighborhood for structurally similar candidate patches using a simple guide image (e.g., fast bilateral filtering), while retrieving the supervision signal directly from the raw noisy neighbor at the matched coordinates. By matching on a noise-attenuated guide while retrieving raw values from neighboring slices, NGPS constructs local pseudo targets without a learned registration module. Across the evaluated CT and synthetic-Rician MRI settings, NGPS improves fidelity and structure-sensitive metrics. Code is available at https://github.com/cv-cho/NGPS .
