Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Qiu, Zhongwei, Chao, Hanqing, Lin, Tiancheng, Chang, Wanxing, Yang, Zijiang, Jiao, Wenpei, Shen, Yixuan, Zhang, Yunshuo, Yang, Yelin, Liu, Wenbin, Jiang, Hui, Bian, Yun, Yan, Ke, Jin, Dakai, Lu, Le
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2412.16711
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866929644363579392
author Qiu, Zhongwei
Chao, Hanqing
Lin, Tiancheng
Chang, Wanxing
Yang, Zijiang
Jiao, Wenpei
Shen, Yixuan
Zhang, Yunshuo
Yang, Yelin
Liu, Wenbin
Jiang, Hui
Bian, Yun
Yan, Ke
Jin, Dakai
Lu, Le
author_facet Qiu, Zhongwei
Chao, Hanqing
Lin, Tiancheng
Chang, Wanxing
Yang, Zijiang
Jiao, Wenpei
Shen, Yixuan
Zhang, Yunshuo
Yang, Yelin
Liu, Wenbin
Jiang, Hui
Bian, Yun
Yan, Ke
Jin, Dakai
Lu, Le
contents Histopathology plays a critical role in medical diagnostics, with whole slide images (WSIs) offering valuable insights that directly influence clinical decision-making. However, the large size and complexity of WSIs may pose significant challenges for deep learning models, in both computational efficiency and effective representation learning. In this work, we introduce Pixel-Mamba, a novel deep learning architecture designed to efficiently handle gigapixel WSIs. Pixel-Mamba leverages the Mamba module, a state-space model (SSM) with linear memory complexity, and incorporates local inductive biases through progressively expanding tokens, akin to convolutional neural networks. This enables Pixel-Mamba to hierarchically combine both local and global information while efficiently addressing computational challenges. Remarkably, Pixel-Mamba achieves or even surpasses the quantitative performance of state-of-the-art (SOTA) foundation models that were pretrained on millions of WSIs or WSI-text pairs, in a range of tumor staging and survival analysis tasks, {\bf even without requiring any pathology-specific pretraining}. Extensive experiments demonstrate the efficacy of Pixel-Mamba as a powerful and efficient framework for end-to-end WSI analysis.
format Preprint
id arxiv_https___arxiv_org_abs_2412_16711
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle From Pixels to Gigapixels: Bridging Local Inductive Bias and Long-Range Dependencies with Pixel-Mamba
Qiu, Zhongwei
Chao, Hanqing
Lin, Tiancheng
Chang, Wanxing
Yang, Zijiang
Jiao, Wenpei
Shen, Yixuan
Zhang, Yunshuo
Yang, Yelin
Liu, Wenbin
Jiang, Hui
Bian, Yun
Yan, Ke
Jin, Dakai
Lu, Le
Computer Vision and Pattern Recognition
Histopathology plays a critical role in medical diagnostics, with whole slide images (WSIs) offering valuable insights that directly influence clinical decision-making. However, the large size and complexity of WSIs may pose significant challenges for deep learning models, in both computational efficiency and effective representation learning. In this work, we introduce Pixel-Mamba, a novel deep learning architecture designed to efficiently handle gigapixel WSIs. Pixel-Mamba leverages the Mamba module, a state-space model (SSM) with linear memory complexity, and incorporates local inductive biases through progressively expanding tokens, akin to convolutional neural networks. This enables Pixel-Mamba to hierarchically combine both local and global information while efficiently addressing computational challenges. Remarkably, Pixel-Mamba achieves or even surpasses the quantitative performance of state-of-the-art (SOTA) foundation models that were pretrained on millions of WSIs or WSI-text pairs, in a range of tumor staging and survival analysis tasks, {\bf even without requiring any pathology-specific pretraining}. Extensive experiments demonstrate the efficacy of Pixel-Mamba as a powerful and efficient framework for end-to-end WSI analysis.
title From Pixels to Gigapixels: Bridging Local Inductive Bias and Long-Range Dependencies with Pixel-Mamba
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2412.16711