Patch-to-Global: 用于全切片图像中全局一致百万像素伪影修复的随机补丁扩散
Patch-to-Global: Random Patch Diffusion for Globally Consistent Megapixel Artifact Inpainting in Whole Slide Images
浏览论文内容
中文总结 AI 辅助
RestorePath框架利用潜在扩散模型和病理基础模型嵌入，通过随机补丁扩散实现全切片图像中百万像素伪影的全局一致修复，提升下游计算病理任务性能。
中文摘要 AI 辅助
尽管深度学习已推进全切片图像（WSI）分析，但气泡和折叠等组织伪影常通过掩盖关键形态导致静默失败。当前病理图像修复方法大多局限于小补丁，难以在百万像素尺度上维持全局结构连贯性。我们提出RestorePath，一个用于全局一致百万像素尺度修复的框架，重建组织学图像中的诊断结构，以防止错误的高置信度预测并降低错误率。我们的模型利用以病理基础模型（PFM）嵌入为条件的潜在扩散模型（LDM），集成大核注意力（LKA）以在随机补丁扩散期间管理长距离依赖。通过距离加权插值（DWI）和自适应引导尺度（AGS）增强，RestorePath通过调节周围补丁的信息确保结构一致性和保真度。在TCGA-BRCA、BACH和Camelyon16数据集上对512至4608像素范围的图像进行评估，展示了在维持组织学一致性方面的最先进性能。RestorePath显著改善下游计算病理学（CP）任务，优于原始伪影图像和传统的检测并丢弃（D&D）方法。代码可在以下https URL获取。
英文摘要
Although deep learning has advanced Whole Slide Image (WSI) Analysis, tissue artifacts like bubbles and folds often cause silent failures by concealing essential morphology. Current pathology image restoration methods are mostly restricted to small patches, struggling to maintain global structural coherence at a megapixel scale. We introduce RestorePath, a framework for globally consistent megapixel scale inpainting that reconstructs diagnostic structures in histological image to prevent incorrect high-confidence predictions and lower error rates. Our model utilizes a Latent Diffusion Model (LDM) conditioned on Pathology Foundation Model (PFM) embeddings, integrating Large Kernel Attention (LKA) to manage long-range dependencies during random patch diffusion. Enhanced by Distance-Weighted Interpolation (DWI) and an Adaptive Guidance Scale (AGS), RestorePath ensures structural consistency and fidelity by modulating information from surrounding patches. Evaluations across TCGA-BRCA, BACH, and Camelyon16 datasets for images ranging from 512 to 4608 pixels demonstrate state-of-the-art performance in maintaining histological consistency. RestorePath significantly improves downstream Computational Pathology (CP) tasks, outperforming both raw artifact images and the conventional Detect-and-Discard (D&D) approach. The code is available at https://github.com/PathfinderLab/RestorePath
发表机构
- Korea University College of Medicine（高丽大学医学院）
- Korea University Anam Hospital（高丽大学安岩医院）
- Seegene Medical Foundation（Seegene医学财团）
- Seoul National University（首尔大学）
- Ewha Womans University（梨花女子大学）
- The Catholic University of Korea（韩国天主教大学）
机构由 AI 辅助整理，请以论文原文为准。
