arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2610.14261cs.CVcs.LG
收藏

BigCell:生成千兆像素级全切片图像

BigCell: Generating Gigapixel Whole-Slide Images

Stony Brook University(石溪大学);Argonne National Laboratory(阿贡国家实验室);University of Chicago(芝加哥大学);The University of Utah(犹他大学);The University of North Carolina at Chapel Hill(北卡罗来纳大学教堂山分校)
Srikar Yellapragada, Alexandros Graikos, Zilinghan Li, Kostas Triaridis, Tarak Nath Nandi, Ji Dong K Bai, Beatrice Knudsen, Tahsin Kurc, Prateek Prasanna, Rajar… 展开作者

Srikar Yellapragada, Alexandros Graikos, Zilinghan Li, Kostas Triaridis, Tarak Nath Nandi, Ji Dong K Bai, Beatrice Knudsen, Tahsin Kurc, Prateek Prasanna, Rajarsi R. Gupta, Ravi K Madduri, Joel Saltz, Dimitris Samaras

AI总结:

BigCell是首个生成千兆像素级全切片组织病理学图像的方法,通过流匹配Transformer等技术实现,其合成样本可提升切片级少样本分类性能最高18%。

摘要:

全切片图像(WSI)是组织病理学组织的千兆像素级数字扫描,是癌症诊断的主要依据。现有组织病理学生成模型在切片尺度运行,生成的固定尺寸图块最大为1024×1024像素,仅覆盖WSI的一小部分。相比之下,临床诊断的文本报告和切片级诊断标签(大规模存在的监督信号)在全切片层面运行。我们提出BigCell,这是首个以原生千兆像素分辨率生成连贯全切片组织病理学图像的方法。BigCell在超过12万个涵盖多种器官类型的公开WSI上,以2048×2048像素训练流匹配Transformer,基于病理学基础模型的嵌入进行条件生成,并采用推理时滑动窗口流水线,生成最大达32768×32768像素的连贯图像,在图块级和切片级保真度指标上均优于现有图块生成器。为使千兆像素合成实用化,我们引入自适应调度策略,优先处理重要切片区域,在质量下降极小的情况下将图像生成速度提升3倍以上。此外,我们提出首个8k分辨率的文本到图像合成方法,通过训练单独模型直接从病理学文本报告生成密集条件网格。我们证明,用合成样本增强数据可使切片级少样本分类性能较基于真实数据训练的基准提升最高18%。

英文摘要:

Whole-slide images (WSIs) are gigapixel-scale digital scans of histopathology tissue and serve as the primary basis for cancer diagnosis. Existing histopathology generative models operate at the patch scale, producing fixed-size tiles of at most 1024$\times$1024 pixels which encompass only a small fraction of the WSI. In contrast, text reports for clinical diagnoses and slide-level diagnostic labels -- the supervision that exists at scale -- operate at the whole-slide level. We present BigCell, the first approach to generate coherent whole-slide histopathology images at native gigapixel resolution. BigCell trains a flow-matching transformer at 2048$\times$2048 pixels on over 120,000 publicly available WSIs spanning multiple organ types. Conditioning on embeddings from pathology foundation models and using an inference-time sliding-window pipeline, BigCell generates coherent images of up to 32,768$\times$32,768 pixels, outperforming prior tile generators on both tile-level and slide-level fidelity metrics. To make gigapixel synthesis practical, we introduce an adaptive scheduling strategy that prioritizes important slide regions, speeding up image generation by over 3\ttimes with minimal quality degradation. Furthermore, we propose the first text-to-image synthesis at 8k resolution by training a separate model that produces dense conditioning grids directly from pathology text reports. We demonstrate that augmenting with synthetic samples improves slide-level few-shot classification performance by up to $18 \%$ over baselines trained on real data.

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。

↑