Polaffini: A feature-based approach for robust affine and polyaffine image registration

2026-02-19

AI Summary

This work proposes an efficient, anatomically plausible closed-form multi-affine registration method that overcomes the limitations of conventional intensity-based approaches—lacking explicit anatomical correspondence—and feature-based methods—hindered by challenging feature extraction. By leveraging a pretrained deep segmentation model to delineate fine-grained anatomical structures, the method establishes one-to-one point correspondences via regional centroids, enabling the integration of global and local affine transformations. Diffeomorphic guarantees are preserved within a log-Euclidean framework, ensuring topological consistency. This approach is the first to combine deep segmentation with centroid-derived features to achieve multi-affine transformations with controllable smoothness. It outperforms state-of-the-art intensity-driven methods in structural alignment accuracy, provides superior initialization for subsequent nonlinear registration, and achieves high precision, robustness, and computational efficiency.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Learning with Manifolds
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

In this work we present Polaffini, a robust and versatile framework for anatomically grounded registration. Medical image registration is dominated by intensity-based registration methods that rely on surrogate measures of alignment quality. In contrast, feature-based approaches that operate by identifying explicit anatomical correspondences, while more desirable in theory, have largely fallen out of favor due to the challenges of reliably extracting features. However, such challenges are now significantly overcome thanks to recent advances in deep learning, which provide pre-trained segmentation models capable of instantly delivering reliable, fine-grained anatomical delineations. We aim to demonstrate that these advances can be leveraged to create new anatomically-grounded image registration algorithms. To this end, we propose Polaffini, which obtains, from these segmented regions, anatomically grounded feature points with 1-to-1 correspondence in a particularly simple way: extracting their centroids. These enable efficient global and local affine matching via closed-form solutions. Those are used to produce an overall transformation ranging from affine to polyaffine with tunable smoothness. Polyaffine transformations can have many more degrees of freedom than affine ones allowing for finer alignment, and their embedding in the log-Euclidean framework ensures diffeomorphic properties. Polaffini has applications both for standalone registration and as pre-alignment for subsequent non-linear registration, and we evaluate it against popular intensity-based registration techniques. Results demonstrate that Polaffini outperforms competing methods in terms of structural alignment and provides improved initialisation for downstream non-linear registration. Polaffini is fast, robust, and accurate, making it particularly well-suited for integration into medical image processing pipelines.

Problem

• image registration
• feature-based
• anatomical correspondence
• affine transformation
• polyaffine transformation

Innovation

• feature-based registration
• polyaffine transformation
• anatomical correspondence
• deep learning segmentation
• log-Euclidean framework