PathVQ：通过向量量化重构计算病理基础模型以用于全切片图像分析
PathVQ: Reforming Computational Pathology Foundation Model for Whole Slide Image Analysis via Vector Quantization
AI总结：
PathVQ通过向量量化蒸馏将病理切片patch token维度从1024压缩至16，并结合多尺度VQ自监督预训练与渐进式卷积模块，在保持重建保真度的同时实现高效的全切片图像分析，取得最先进性能。
摘要：
计算病理学与全切片图像（WSI）分析在癌症诊断和预后中至关重要。然而，WSI的超高分辨率带来了显著的建模挑战。近年来病理基础模型的进展已提升了性能，但大多数方法依赖图块ViT的[CLS]标记表示作为切片级输入（16x16像素称为patch，224x224像素称为tile）。这丢弃了来自patch token的关键空间细节，限制了下游WSI分析任务。我们发现利用所有空间patch token有利于WSI分析，但会带来近200倍的存储和训练成本增加（例如，ViT$_{224}$中的196个token）。为解决这一问题，我们引入了基于patch特征的向量量化（VQ）蒸馏，利用离散索引和解码器高效压缩空间patch token。我们的方法将token维度从1024降至16，实现了64倍压缩率，同时保持重建保真度。此外，我们采用多尺度VQ（MSVQ）策略，不仅提升了VQ重建性能，还作为自监督学习（SSL）监督信号，实现无缝的切片级预训练目标。基于经MSVQ量化的patch特征和tile监督目标，我们开发了渐进式卷积模块和切片级SSL，以提取具有丰富空间信息的表示用于下游WSI任务。在多个数据集上的广泛评估证明了我们方法的有效性，在WSI分析中取得了最先进的性能。代码即将发布。
英文摘要：
Computational pathology and whole-slide image (WSI) analysis are pivotal in cancer diagnosis and prognosis. However, the ultra-high resolution of WSIs presents significant modeling challenges. Recent advancements in pathology foundation models have improved performance, yet most approaches rely on [CLS] token representation of tile ViT as slide-level inputs (16x16 pixels is refereed as patch and 224x224 pixels as tile). This discards critical spatial details from patch tokens, limiting downstream WSI analysis tasks. We find that leveraging all spatial patch tokens benefits WSI analysis but incurs nearly 200x higher storage and training costs (e.g., 196 tokens in ViT$_{224}$). To address this, we introduce vector quantized (VQ) distillation on patch feature, which efficiently compresses spatial patch tokens using discrete indices and a decoder. Our method reduces token dimensionality from 1024 to 16, achieving a 64x compression rate while preserving reconstruction fidelity. Furthermore, we employ a multi-scale VQ (MSVQ) strategy, which not only enhances VQ reconstruction performance but also serves as a Self-supervised Learning (SSL) supervision for a seamless slide-level pretraining objective. Built upon the quantized patch features and supervision targets of tile via MSVQ, we develop a progressive convolutional module and slide-level SSL to extract representations with rich spatial-information for downstream WSI tasks. Extensive evaluations on multiple datasets demonstrate the effectiveness of our approach, achieving state-of-the-art performance in WSI analysis. Code will be available soon.
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。
