LayersReg: A Layer-by-Layer Progressive Regressor for Reliable Intraoperative 3D/2D Registration

2026-06-25

AI Summary

This work addresses the limitations of conventional iterative methods—such as low efficiency and high failure rates—and the insufficient generalization of existing deep learning models in intraoperative 3D/2D registration. To overcome these challenges, the authors propose a hierarchical, progressive regression framework endowed with 3D anatomical awareness. The approach formulates registration as a layer-wise pose estimation process within a progressively refined search space, integrating node-level regression, cross-modality image feature correspondence, and pixel-flow trends to achieve anatomy-aware and precise alignment. Evaluated under large initial misalignments and multimodal settings, the method achieves state-of-the-art accuracy of 0.68° and 1.41 mm for X-ray/CT registration and 0.73° and 1.55 mm for slice localization, substantially outperforming current best practices and meeting the stringent demands of real-time, high-precision intraoperative guidance.

Abstract

3D/2D registration serves as a cornerstone technique in surgical navigation. Traditional iterative optimization algorithms suffer from low efficiency and high failure rates in intraoperative settings. Deep learning-based methods reformulate registration from iterative optimization to a regression problem that maps image appearance features to spatial pose, typically achieving improved real-time performance and accuracy. However, such learnable methods are confined to memory-driven retrieval of specific pose features rather than understanding the task of image alignment itself, which limits their generalization in complex scenarios. We propose LayersReg, a pioneering regression paradigm that endows the model with 3D anatomical awareness and searches for the correct pose in a progressive, layer-by-layer manner. Inspired by the iterative pose-searching optimization criterion of classical registration, LayersReg searches for correlations between the moving and fixed images in feature space, capturing the trend of pixel flow and thereby converging iteratively toward the correct spatial pose transformation. We further design a coupling of node-wise regression with the progressive registration framework to enhance the model's perception of spatial pose changes. Experimental results demonstrate that under large offsets and multimodality conditions, LayersReg achieves high accuracy on both X-ray/CT registration (0.68°, 1.41 mm) and slice localization (0.73°, 1.55 mm) tasks, outperforming existing state-of-the-art methods while meeting the intraoperative demands for precision and real-time capability.

Problem

• 3D/2D registration
• intraoperative navigation
• deep learning generalization
• multimodality
• pose estimation

Innovation

• layer-by-layer regression
• 3D/2D registration
• progressive pose search
• anatomical awareness
• deep learning-based registration