Promptable Representation Distribution Learning and Data Augmentation for Gigapixel Histopathology WSI Analysis

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Tang, Kunming, Jiang, Zhiguo, Shi, Jun, Wang, Wei, Wu, Haibo, Zheng, Yushan
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912161233633280
author Tang, Kunming
Jiang, Zhiguo
Shi, Jun
Wang, Wei
Wu, Haibo
Zheng, Yushan
author_facet Tang, Kunming
Jiang, Zhiguo
Shi, Jun
Wang, Wei
Wu, Haibo
Zheng, Yushan
contents Gigapixel image analysis, particularly for whole slide images (WSIs), often relies on multiple instance learning (MIL). Under the paradigm of MIL, patch image representations are extracted and then fixed during the training of the MIL classifiers for efficiency consideration. However, the invariance of representations makes it difficult to perform data augmentation for WSI-level model training, which significantly limits the performance of the downstream WSI analysis. The current data augmentation methods for gigapixel images either introduce additional computational costs or result in a loss of semantic information, which is hard to meet the requirements for efficiency and stability needed for WSI model training. In this paper, we propose a Promptable Representation Distribution Learning framework (PRDL) for both patch-level representation learning and WSI-level data augmentation. Meanwhile, we explore the use of prompts to guide data augmentation in feature space, which achieves promptable data augmentation for training robust WSI-level models. The experimental results have demonstrated that the proposed method stably outperforms state-of-the-art methods.
format Preprint
id arxiv_https___arxiv_org_abs_2412_14473
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Promptable Representation Distribution Learning and Data Augmentation for Gigapixel Histopathology WSI Analysis
Tang, Kunming
Jiang, Zhiguo
Shi, Jun
Wang, Wei
Wu, Haibo
Zheng, Yushan
Computer Vision and Pattern Recognition
Gigapixel image analysis, particularly for whole slide images (WSIs), often relies on multiple instance learning (MIL). Under the paradigm of MIL, patch image representations are extracted and then fixed during the training of the MIL classifiers for efficiency consideration. However, the invariance of representations makes it difficult to perform data augmentation for WSI-level model training, which significantly limits the performance of the downstream WSI analysis. The current data augmentation methods for gigapixel images either introduce additional computational costs or result in a loss of semantic information, which is hard to meet the requirements for efficiency and stability needed for WSI model training. In this paper, we propose a Promptable Representation Distribution Learning framework (PRDL) for both patch-level representation learning and WSI-level data augmentation. Meanwhile, we explore the use of prompts to guide data augmentation in feature space, which achieves promptable data augmentation for training robust WSI-level models. The experimental results have demonstrated that the proposed method stably outperforms state-of-the-art methods.
title Promptable Representation Distribution Learning and Data Augmentation for Gigapixel Histopathology WSI Analysis
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2412.14473