一种基于拉普拉斯金字塔的生成式 H&E 染色增强网络
A Laplacian Pyramid Based Generative H&E Stain Augmentation Network
AI总结：
针对 H&E 染色颜色差异影响模型泛化的问题，本文提出基于 GAN 与拉普拉斯金字塔生成器的 G-SAN，以解耦染色和形态并生成真实染色变化，显著提升分类与分割性能。
摘要：
苏木精和伊红（H&E）染色是一种广泛使用的样本制备流程，用于提高组织切片的饱和度以及组织学图像中细胞核与细胞质之间的对比度，以服务于医学诊断。然而，多种因素（例如所用试剂的差异）会导致实际记录的染色颜色具有高度可变性。这种可变性给基于机器学习的计算机辅助诊断工具实现泛化带来了挑战。为了降低所学模型对染色变化的敏感性，我们提出了生成式染色增强网络（Generative Stain Augmentation Network，G-SAN）——这是一个基于 GAN 的框架，利用模拟但真实的染色变化来增强细胞图像集合。G-SAN 的核心采用了一种新颖且计算效率很高的基于拉普拉斯金字塔（Laplacian Pyramid，LP）的生成器架构，该架构能够将染色与细胞形态解耦。通过图像块分类和细胞核分割任务，我们表明，使用经 G-SAN 增强的训练数据分别使 F1 分数平均提高 15.7%，使全景质量平均提高 7.3%。我们的代码可在 https://github.com/lifangda01/GSAN-Demo 获取。
英文摘要：
Hematoxylin and Eosin (H&E) staining is a widely used sample preparation procedure for enhancing the saturation of tissue sections and the contrast between nuclei and cytoplasm in histology images for medical diagnostics. However, various factors, such as the differences in the reagents used, result in high variability in the colors of the stains actually recorded. This variability poses a challenge in achieving generalization for machine-learning based computer-aided diagnostic tools. To desensitize the learned models to stain variations, we propose the Generative Stain Augmentation Network (G-SAN) -- a GAN-based framework that augments a collection of cell images with simulated yet realistic stain variations. At its core, G-SAN uses a novel and highly computationally efficient Laplacian Pyramid (LP) based generator architecture, that is capable of disentangling stain from cell morphology. Through the task of patch classification and nucleus segmentation, we show that using G-SAN-augmented training data provides on average 15.7% improvement in F1 score and 7.3% improvement in panoptic quality, respectively. Our code is available at https://github.com/lifangda01/GSAN-Demo.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
