Abstract

Non-rigid medical image registration (NRR) is critical for aligning multi-modal anatomical structures but is challenged by complex physiological deformations (e.g., respiration, contrast variations). Traditional NRR incurs high computational costs and requires extensive parameter tuning, while deep learning (DL) approaches rely on restrictive task-specific retraining. Addressing these limitations, we introduce a novel AI-driven 3D NRR framework that generalizes across diverse imaging modalities and anatomies. By eliminating anatomy- and modality-specific customization, our approach provides an efficient, universal solution for streamlined integration into heterogeneous clinical workflows.

Links to Paper and Supplementary Materials

Main Paper (Open Access Version): https://papers.miccai.org/miccai-2026-sat/paper/MLMI_018.pdf

SharedIt Link: Not yet available

SpringerLink (DOI): Not yet available

Supplementary Material: Not Submitted

Link to Open Review

Open Review Page: https://openreview.net/forum?id=dJbwhDlXH0

BibTex

@InProceedings{CSne_Efficient_MICCAISAT2026,
        author = { C, Sneha Sree AND Shanbhag, Dattesh D. AND Chatterjee, Sudhanya},
        title = { { Efficient Test-Time Optimization via Knowledge Distillation for Modality-Agnostic Medical Image Registration } },
        booktitle = {Medical Image Computing and Computer Assisted Intervention -- MICCAI 2026 Workshops and Challenges},
        year = {2026},
        publisher = {Springer Nature Switzerland},
        volume = {LNCS 17272},
        month = {pending},
        page = {pending}
}


back to top