Learning from the Right Patches: A Two-Stage Wavelet-Driven Masked Autoencoder for Histopathology Representation Learning

2025/11/10 by Raneen Younis, Younis, Raneen, Louay Hamdi +5
Computer Science · #AI in cancer detection #Computer Vision and Pattern Recognition (cs.CV) #Digital Imaging for Blood Diseases #FOS: Computer and information sciences #Generative Adversarial Networks and Image Synthesis

paper · pdf · doi:10.48550/arxiv.2511.06958

openalex publication_date 2025/11/10 · openalex created_date 2025/11/12 · openalex updated_date 2026/07/28

Abstract

Whole-slide images are central to digital pathology, yet their extreme size and scarce annotations make self-supervised learning essential. Masked Autoencoders (MAEs) with Vision Transformer backbones have recently shown strong potential for histopathology representation learning. However, conventional random patch sampling during MAE pretraining often includes irrelevant or noisy regions, limiting the model's ability to capture meaningful tissue patterns. In this paper, we present a lightweight and domain-adapted framework that brings structure and biological relevance into MAE-based learning through a wavelet-informed patch selection strategy. WISE-MAE applies a two-step coarse-to-fine process: wavelet-based screening at low magnification to locate structurally rich regions, followed by high-resolution extraction for detailed modeling. This approach mirrors the diagnostic workflow of pathologists and improves the quality of learned representations. Evaluations across multiple cancer datasets, including lung, renal, and colorectal tissues, show that WISE-MAE achieves competitive representation quality and downstream classification performance while maintaining efficiency under weak supervision.

Citations

• Multi-head Attention-based Deep Multiple Instance Learning
• Dynamic Graph Representation with Knowledge-aware Attention for Histopathology Whole Slide Image Analysis
• FreMIM: Fourier Transform Meets Masked Image Modeling for Medical Image Segmentation
• Diagnose Like a Pathologist: Transformer-Enabled Hierarchical Attention-Guided Multiple Instance Learning for Whole Slide Image Classification
• Global Contrast Masked Autoencoders Are Powerful Pathological Representation Learners
• DTFD-MIL: Double-Tier Feature Distillation Multiple Instance Learning for Histopathology Whole Slide Image Classification
• Masked Feature Prediction for Self-Supervised Visual Pre-Training
• Masked Autoencoders Are Scalable Vision Learners
• Self-Supervised Representation Learning using Visual Field Expansion on\n Digital Pathology
• TransMIL: Transformer based Correlated Multiple Instance Learning for Whole Slide Image Classification
• Cluster-to-Conquer: A Framework for End-to-End Multi-Instance Learning for Whole Slide Image Classification
• Dual-stream Multiple Instance Learning Network for Whole Slide Image Classification with Self-supervised Contrastive Learning
• A Multi-resolution Model for Histopathology Image Classification and Localization with Multiple Instance Learning
• A multi-resolution model for histopathology image classification and localization with multiple instance learning
• An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
• Contrastive Learning of Medical Visual Representations from Paired Images and Text
• SelfAugment: Automatic Augmentation Policies for Self-Supervised\n Learning
• Data Efficient and Weakly Supervised Computational Pathology on Whole Slide Images
• Multi-Resolution A*
• Multi-scale Domain-adversarial Multiple-instance CNN for Cancer Subtype Classification with Unannotated Histopathological Images
• Momentum Contrast for Unsupervised Visual Representation Learning
• CAMEL: A Weakly Supervised Learning Framework for Histopathology Image Segmentation
• Adaptive Weighting Multi-Field-of-View CNN for Semantic Segmentation in Pathology
• Patch Clustering for Representation of Histopathology Images
• Unsupervised Representation Learning by Predicting Image Rotations
• Attention-based Deep Multiple Instance Learning
• Attention Is All You Need
• Patch-based Convolutional Neural Network for Whole Slide Tissue Image Classification

Related

• Multi-modal Masked Autoencoders Learn Compositional Histopathological Representations
• Pathology Image Compression with Pre-trained Autoencoders
• Self-distillation Augmented Masked Autoencoders for Histopathological Image Classification
• Computational Pathology at Health System Scale -- Self-Supervised Foundation Models from Three Billion Images
• Medical supervised masked autoencoders: Crafting a better masking strategy and efficient fine-tuning schedule for medical image classification