Adaptive Riemannian optimization for multi-scale diffeomorphic matching

Clinical Snapshot

Key Findings

• P Value: Not reported

• Effect Size: 2x faster than ANTs on CPU, 100x faster on GPU

• Primary Outcome: Computational performance improvement in image registration

• Nnt Or Sensitivity: 10x reduction in memory consumption compared to deep learning methods

• Confidence Interval: Not reported

Clinical Application

High - training-free algorithm reduces implementation barriers, GPU acceleration widely available in research institutions Highly relevant to Australian medical imaging research infrastructure, particularly for multi-centre studies and resource-constrained environments. Could enhance efficiency of imaging biomarker research supported by NHMRC funding Medical imaging researchers and clinicians requiring image registration across modalities, longitudinal studies, and multi-subject analyses

Abstract

Image matching is a fundamental task in quantitative biomedical and biological image analyses, enabling researchers to compare, integrate, and interpret imaging data across subjects, time points, modalities, and experimental conditions. Existing state-of-the-art registration methods are slow due to inefficient implementations and poor convergence rates because of the ill-conditioned nature of the optimization problem. Deep learning methods offer fast inference but require extensive training time, substantial inference memory, and fail to generalize across long-tailed distributions or diverse image modalities, necessitating costly retraining. We address these challenges by proposing FireANTs, a training-free, GPU-accelerated, multi-scale adaptive Riemannian optimization algorithm for fast and accurate dense diffeomorphic image matching. FireANTs more than doubles the speed of the community standard ANTs registration tool on a CPU, and is two orders of magnitude faster on a GPU. On the GPU, FireANTs performs competitively with deep learning methods on inference runtime while consuming up to 10 × less memory. FireANTs demonstrates robustness on a wide variety of matching problems across modalities, species, and organs, without any domain-specific training or tuning. Our framework allows hyperparameter grid search studies with less resources and time compared to traditional and deep learning registration algorithms alike.

References

1. 1.Jena, R., Chaudhari, P., & Gee, J. C. (2026). Adaptive Riemannian optimization for multi-scale diffeomorphic matching. Nature Communications. https://doi.org/10.1007/s00039-006-0573-8