eess.IVAug 27, 2026

Is Deformable Image Registration Ready for Brain Metastasis Reirradiation Dose Accumulation? A Longitudinal MRI Benchmark of Registration Accuracy

Authors: Hengjie Liu, Manju Sharma, Xinyi Fu, Di Xu, Ke Sheng

Organizations: Department of Radiation Oncology, University of California, San Francisco, San Francisco, CA, USA

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Dose accumulation is increasingly important in adaptive radiation therapy and reirradiation, but its clinical validity depends on the performance of deformable image registration (DIR). Reirradiation of brain metastases (BMs) with stereotactic radiosurgery (SRS) provides a controlled but clinically meaningful DIR test case: intra-subject brain deformation is usually limited after rigid alignment, yet recurrent lesions can undergo substantial local shape and volume changes that rigid registration cannot capture and can affect dose accumulation. We benchmarked a wide range of learning-based and optimization-based DIR methods on 87 manually screened longitudinal contrast-enhanced T1-weighted MRI lesion pairs from an institutional BM SRS retreatment cohort. Learning-based methods pretrained on healthy-brain MRI were evaluated zero-shot and after instance-specific optimization (ISO) or tumor-proximity target-specific optimization (TSO). Registration was assessed using lesion overlap (Dice), surface distance metrics (HD95 and sASD), target-volume recovery, and runtime and memory. Pretrained learning-based methods showed variable zero-shot performance, while ISO/TSO improved all tested learning-based families. However, optimization-based methods remained the best-performing approach while maintaining reasonable runtime. These findings suggest that even state-of-the-art DIR methods do not yet provide sufficiently accurate and consistent registration for unmonitored use in brain metastasis reirradiation dose accumulation. Because accurate registration is a prerequisite for deformable dose accumulation, clinical application will require case-level quality control and direct assessment of how registration uncertainty affects downstream dose metrics.

Figures & tables

Figure 1: Representative included and excluded lesion pairs. Red contours show lesion masks on initial and repeat SRS MRI; insets show local dose overlays. Excluded cases had poor intensity correspondence for DIR.
Figure 2: Registration accuracy and target-volume recovery. (A-B) Dice and HD95 for rigid, learning-based, and optimization-based methods. (C-J) Recovered versus observed log2 target-volume change for selected methods; dashed lines indicate ideal recovery.
MethodDiceHD95sASDVol. slopeVol. rr
Baseline
Rigid0.417 (0.173)4.62 (3.10)2.23 (1.53)––
Learning-based
VM-PT0.582 (0.172)3.44 (2.81)1.03 (0.81)0.3840.786
VM-ISO0.622 (0.163)3.07 (2.54)0.88 (0.68)0.5150.873
VM-TSO0.640 (0.174)3.02 (2.75)0.83 (0.71)0.5600.866
Table 1: Registration accuracy and target-volume agreement on the 87-case cohort. Dice, HD95, and sASD are reported as mean (standard deviation). HD95 and sASD are in mm. Volume agreement reports the regression slope and Pearson correlation between warped and fixed target volumes. Bold indicates the top three methods for each metric; HD95/sASD are ranked lower-is-better, slope by closeness to 1, and rr higher-is-better.
MethodPTISO/TSO or optimization
VoxelMorph0.23 s / 2.9 GB94 s / 13 GB
TransMorph0.30 s / 1.4 GB124 s / 9.4 GB
VFA0.69 s / 8.7 GB186 s / 40 GB
SITReg1.33 s / 4.9 GB410 s / 28 GB
UniGradICON0.62 s / 2.7 GB203 s / 12 GB
Greedy (CPU)–344 s / –
Table 2: Runtime and peak GPU memory. Values are reported as mean runtime and peak memory. CPU-only methods do not report GPU memory.
More in this topic

Deformable Image Registration

Deformable image registration aligns images by estimating local spatial transformations that account for anatomical change, motion, and differences between scans.

See all 77 papers

Explore similar work

CardsList
Oct 13, 2026

Personalising Deformable Image Registration with Clinical Context and Deformation History for Adaptive Radiotherapy

Adaptive radiotherapy must account for anatomical changes during treatment, but doing so depends on reliably mapping targets, organs at risk and dose across serial images. Existing registration methods treat each image pair independently, without using radiotherapy-specific context or the patient's preceding anatomical trajectory. We developed Clinical-context-Aware and Patient-adaptive Deformable Image Registration (CAP-DIR), integrating source contours, dose, clinical text and prior-registration features. We evaluated 16,361 CT examinations from 3,473 patients across four institutional cohorts and five public benchmarks. CAP-DIR increased internal-test organ-at-risk-region normalized cross-correlation from 84.76% to 96.92% relative to the strongest learning-based comparator, with low deformation folding, and remained superior across seven external cohorts. In a blinded review of 20 longitudinal cases, clinically acceptable contours increased from 45.0% to 82.5% for clinical target volumes and from 22.5% to 62.5% for organs at risk, while estimated editing time decreased. In a controlled benchmark, CAP-DIR reduced global dose-mapping error by 34.5% relative to the strongest comparator. These findings show that clinically informed, longitudinal registration can improve contour usability and dose mapping for adaptive radiotherapy.

Caiwen Jiang, Songlin Zhao, Yuzhen Ding +23

Department of Radiation Oncology, Mayo Clinic, Phoenix, Arizona, USA · School of Biomedical Engineering, ShanghaiTech University, Shanghai, China · Department of Radiation Oncology, Mayo Clinic, Rochester, Minnesota, USA +2

Medical Image RegistrationDeformable Image Registration
AskPDF
Oct 16, 2026

LUMINR: Intensity-conditioned Meta-learned INRs for Accelerated Medical Image Registration

Implicit neural representations (INRs) have been applied to deformable image registration, where a lightweight network maps a spatial coordinate to a deformation vector. Because an INR encodes one specific signal in its parameters, a fresh network must be optimized for every image pair at test time, which is slow. Existing methods reduce this cost by meta-learning an initialization across a population of image pairs, and condition the weights of the INR globally, so the same conditioning applies at every coordinate while the deformation field it initializes varies across the domain. We propose LUMINR, a meta-learning framework that accelerates this per-pair fitting and in which the conditioning is local instead. LUMINR uses an INR that maps a coordinate, together with local intensity and gradient features, to a deformation vector, so the conditioning varies over the domain at the resolution of the field being modelled. Its objective is to find an initialization of the INR parameters from which the network can be fine-tuned to register any unseen pair in only a few optimization iterations. We evaluate LUMINR on 3D deformable registration of brain MRI from the OASIS dataset, where local conditioning and the meta-learned initialization prove complementary: early in fitting they reach a markedly higher accuracy than either does alone. Fine-tuned from that initialization, LUMINR matches the accuracy of the same conditioned INR fit from scratch in roughly a tenth of the gradient steps and a tenth of the wall-clock time, while the deformation fields stay smooth with negligible folding.

Damian Rynczak, Marek Wodzinski

AGH University of Krakow · Sano - Centre for Computational Personalised Medicine, International Research Foundation

Medical Image RegistrationDeformable Image RegistrationImplicit Neural RepresentationsMeta-Learning
AskPDF
Jun 8, 2026

A practical probabilistic framework for deformable image registration uncertainty in radiotherapy dose propagation

Deformable image registration (DIR) is widely used in radiotherapy for dose propagation and accumulation, but uncertainty in the underlying deformation can substantially affect clinically relevant dose estimates. We present a practical probabilistic framework for propagating DIR uncertainty to voxel-wise dose statistics and dose-volume histograms (DVHs). The method models the mapped correspondence at each voxel as a random variable governed by a transparent local certainty map that can be defined by simple safety margins, structure-boundary mismatch, or structure-wise conservative uncertainty values. This yields interpretable quantities such as dose probabilities, expected dose, confidence bounds, and induced DVH envelopes. The framework is designed to remain lightweight and interpretable: it avoids complex biomechanical or ensemble-based uncertainty models and instead emphasizes simple parameterization, computational feasibility, and transparent dose metrics. We further introduce a structure-guided in/out strategy as an optional refinement that restricts mapping probabilities to anatomically plausible target regions. The approach is demonstrated on a prostate radiotherapy case study and used to compare different certainty-map strategies and probability kernels. The experiments show that the certainty-map design has a stronger effect on resulting dose and DVH uncertainty bounds than the specific kernel choice, while the additional benefit of the in/out strategy is case-dependent and modest in the present example. Overall, the proposed framework provides a transparent way to incorporate DIR uncertainty into radiotherapy dose assessment and to study how modelling choices affect propagated dose metrics.

Stefan Heldmann, Sven Kuckertz, Nasim Givehchi +4

Fraunhofer Institute for Digital Medicine MEVIS, Lübeck, Germany · Varian, a Siemens Healthineers company, Dättwil, Switzerland · Icon Group, South Brisbane, QLD, Australia

Uncertainty QuantificationOncologyUncertainty PropagationDeformable Image Registration
AskPDF