CORE - A Cell-Level Coarse-to-Fine Image Registration Engine for Multi-stain Image Alignment

2025-11-05

AI Summary

This work addresses the low accuracy and poor robustness of nucleus-level cross-modal (e.g., brightfield/fluorescence) registration for multi-stained whole-slide images (WSIs). Methodologically, we propose a “coarse-to-fine” three-stage registration engine: (1) prompt-guided tissue mask segmentation for coarse global alignment; (2) self-supervised nuclear center point detection to construct morphology-aware point sets; and (3) fusion of pretrained dense feature matching with Coherent Point Drift (CPD) to optimize nonlinear displacement fields. Our key contribution lies in the organic integration of tissue structural priors, nuclear geometric constraints, and deep feature matching. Evaluated on three public and two private datasets, our method significantly outperforms state-of-the-art approaches in registration accuracy, generalizability, and computational efficiency—achieving, for the first time, highly robust and precise nucleus-level cross-modal alignment.

Technology Category

• Computer Vision: Multi-modal Vision
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Multimodal Learning

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Accurate and efficient registration of whole slide images (WSIs) is essential for high-resolution, nuclei-level analysis in multi-stained tissue slides. We propose a novel coarse-to-fine framework CORE for accurate nuclei-level registration across diverse multimodal whole-slide image (WSI) datasets. The coarse registration stage leverages prompt-based tissue mask extraction to effectively filter out artefacts and non-tissue regions, followed by global alignment using tissue morphology and ac- celerated dense feature matching with a pre-trained feature extractor. From the coarsely aligned slides, nuclei centroids are detected and subjected to fine-grained rigid registration using a custom, shape-aware point-set registration model. Finally, non-rigid alignment at the cellular level is achieved by estimating a non-linear dis- placement field using Coherent Point Drift (CPD). Our approach benefits from automatically generated nuclei that enhance the accuracy of deformable registra- tion and ensure precise nuclei-level correspondence across modalities. The pro- posed model is evaluated on three publicly available WSI registration datasets, and two private datasets. We show that CORE outperforms current state-of-the-art methods in terms of generalisability, precision, and robustness in bright-field and immunofluorescence microscopy WSIs

Problem

• Accurate nuclei-level registration for multi-stained whole slide images
• Aligning cellular structures across bright-field and immunofluorescence microscopy
• Establishing precise nuclei correspondence in multimodal WSI datasets

Innovation

• Coarse registration uses prompt-based tissue mask extraction
• Fine-grained rigid registration employs shape-aware point-set model
• Non-rigid cellular alignment applies Coherent Point Drift technique