Robust Multi-Scale Implicit Neural Representations for Large-Deformation Lung Registration

Johannes B. Gebauer, Maximilian Nielsen, Frederic Madesta, René Werner, Thilo Sentker
Proceedings of The 9th International Conference on Medical Imaging with Deep Learning, PMLR 315:3089-3102, 2026.

Abstract

We propose a multi-scale Implicit Neural Representation (INR) framework for dense deformable image registration, designed to stabilize convergence for large deformations while preserving precision for fine anatomical details. We model the INR as a dual-branch architecture that explicitly decomposes the motion into global and local components. The objective function is driven by mask-guided Normalized Cross-Correlation augmented by geometric and semantic regularization to ensure smooth, anatomically plausible motion. Evaluation on the DIR-Lab 4DCT thorax dataset demonstrates competitive performance with a mean Target Registration Error (TRE) below 1.0\,mm. On the more challenging DIR-Lab COPDgene thorax dataset, the model achieves robust alignment with a mean TRE of 1.23\,mm, yielding performance comparable to leading classical optimization frameworks. A comprehensive ablation study confirms that the dual-branch design and multi-scale optimization strategy are necessary to achieve these results, enabling stable registration with modest computational overhead.

Cite this Paper

BibTeX
@InProceedings{pmlr-v315-gebauer26a, title = {Robust Multi-Scale Implicit Neural Representations for Large-Deformation Lung Registration}, author = {Gebauer, Johannes B. and Nielsen, Maximilian and Madesta, Frederic and Werner, Ren{\'e} and Sentker, Thilo}, booktitle = {Proceedings of The 9th International Conference on Medical Imaging with Deep Learning}, pages = {3089--3102}, year = {2026}, editor = {Huo, Yuankai and Gao, Mingchen and Kuo, Chang-Fu and Jin, Yueming and Deng, Ruining}, volume = {315}, series = {Proceedings of Machine Learning Research}, month = {08--10 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v315/main/assets/gebauer26a/gebauer26a.pdf}, url = {https://proceedings.mlr.press/v315/gebauer26a.html}, abstract = {We propose a multi-scale Implicit Neural Representation (INR) framework for dense deformable image registration, designed to stabilize convergence for large deformations while preserving precision for fine anatomical details. We model the INR as a dual-branch architecture that explicitly decomposes the motion into global and local components. The objective function is driven by mask-guided Normalized Cross-Correlation augmented by geometric and semantic regularization to ensure smooth, anatomically plausible motion. Evaluation on the DIR-Lab 4DCT thorax dataset demonstrates competitive performance with a mean Target Registration Error (TRE) below 1.0\,mm. On the more challenging DIR-Lab COPDgene thorax dataset, the model achieves robust alignment with a mean TRE of 1.23\,mm, yielding performance comparable to leading classical optimization frameworks. A comprehensive ablation study confirms that the dual-branch design and multi-scale optimization strategy are necessary to achieve these results, enabling stable registration with modest computational overhead.} }
Endnote
%0 Conference Paper %T Robust Multi-Scale Implicit Neural Representations for Large-Deformation Lung Registration %A Johannes B. Gebauer %A Maximilian Nielsen %A Frederic Madesta %A René Werner %A Thilo Sentker %B Proceedings of The 9th International Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2026 %E Yuankai Huo %E Mingchen Gao %E Chang-Fu Kuo %E Yueming Jin %E Ruining Deng %F pmlr-v315-gebauer26a %I PMLR %P 3089--3102 %U https://proceedings.mlr.press/v315/gebauer26a.html %V 315 %X We propose a multi-scale Implicit Neural Representation (INR) framework for dense deformable image registration, designed to stabilize convergence for large deformations while preserving precision for fine anatomical details. We model the INR as a dual-branch architecture that explicitly decomposes the motion into global and local components. The objective function is driven by mask-guided Normalized Cross-Correlation augmented by geometric and semantic regularization to ensure smooth, anatomically plausible motion. Evaluation on the DIR-Lab 4DCT thorax dataset demonstrates competitive performance with a mean Target Registration Error (TRE) below 1.0\,mm. On the more challenging DIR-Lab COPDgene thorax dataset, the model achieves robust alignment with a mean TRE of 1.23\,mm, yielding performance comparable to leading classical optimization frameworks. A comprehensive ablation study confirms that the dual-branch design and multi-scale optimization strategy are necessary to achieve these results, enabling stable registration with modest computational overhead.
APA
Gebauer, J.B., Nielsen, M., Madesta, F., Werner, R. & Sentker, T.. (2026). Robust Multi-Scale Implicit Neural Representations for Large-Deformation Lung Registration. Proceedings of The 9th International Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 315:3089-3102 Available from https://proceedings.mlr.press/v315/gebauer26a.html.

Related Material