From Traditional to Deep Learning Approaches in Whole Slide Image Registration: A Methodological Review

2025-02-26

AI Summary

Whole-slide image (WSI) registration is critical for tumor microenvironment (TME) analysis but faces major challenges including prohibitive computational cost due to ultra-high resolution, inter-stain and cross-layer appearance variations, non-rigid tissue deformations, and artifacts introduced during slide preparation. This paper presents the first systematic survey of WSI registration methodologies, categorizing them into traditional feature-based approaches (e.g., SIFT, ORB), phase-correlation methods, and deep learning paradigms (e.g., U-Net, Transformer, self-supervised architectures), while clarifying their respective applicability domains and performance limits. We unify and catalog publicly available benchmark datasets, standardized evaluation metrics, and open-source software tools. Furthermore, we identify persistent bottlenecks—particularly in robustness, generalizability, and multi-modal alignment—and propose a clinically oriented, scalable, and decoupled technical roadmap for WSI registration. This work fills a critical gap by providing the first comprehensive, taxonomy-driven review of this rapidly evolving field.

Technology Category

• Computer Vision: Medical and Biological Imaging
• Machine Learning: Multi-instance/Multi-view Learning
• Search and Optimization: Metareasoning and Metaheuristics

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Graph Algorithms and Modeling for the Web: Querying, indexing, and retrieval in Web-related graphs

Abstract

Whole slide image (WSI) registration is an essential task for analysing the tumour microenvironment (TME) in histopathology. It involves the alignment of spatial information between WSIs of the same section or serial sections of a tissue sample. The tissue sections are usually stained with single or multiple biomarkers before imaging, and the goal is to identify neighbouring nuclei along the Z-axis for creating a 3D image or identifying subclasses of cells in the TME. This task is considerably more challenging compared to radiology image registration, such as magnetic resonance imaging or computed tomography, due to various factors. These include gigapixel size of images, variations in appearance between differently stained tissues, changes in structure and morphology between non-consecutive sections, and the presence of artefacts, tears, and deformations. Currently, there is a noticeable gap in the literature regarding a review of the current approaches and their limitations, as well as the challenges and opportunities they present. We aim to provide a comprehensive understanding of the available approaches and their application for various purposes. Furthermore, we investigate current deep learning methods used for WSI registration, emphasising their diverse methodologies. We examine the available datasets and explore tools and software employed in the field. Finally, we identify open challenges and potential future trends in this area of research.

Problem

• Whole slide image alignment challenges
• Deep learning in WSI registration
• Tumour microenvironment analysis methods

Innovation

• Deep learning for WSI registration
• Comprehensive methodological review
• Exploration of datasets and tools