Abstract
In computer vision and graphics, 3D shape alignment is a fundamental task for data analysis. Recently, techniques from those fields have been successfully adapted to medical imaging, specifically in Statistical Shape Analysis (SSA), by leveraging non-rigid spectral matching frameworks such as Functional Maps (FMaps). However, current methods typically rely on surface-based representations derived from volumetric modalities such as Magnetic Resonance Imaging (MRI), thereby discarding intrinsic volumetric information. To address this limitation, we shift the paradigm by adopting a matching pipeline based on the newly introduced Volumetric Functional Maps framework (Volumetric FMaps), extending functional correspondence to the volumetric domain. Evaluating this approach on collections of MRI-derived volumetric neuroanatomical meshes, we demonstrate that Volumetric FMaps produce accurate, dense 3D correspondences. Furthermore, we leverage these correspondences to transfer clinically relevant information, such as anatomical segmentations and keypoints, from a template shape to target samples. By showcasing the utility of this framework, this work establishes a first step toward fully incorporating volumetric data into SSA workflows. To foster reproducibility and research in the field code is openly available at: https://github.com/Franca-exe/MedTetMesh
Links to Paper and Supplementary Materials
Main Paper (Open Access Version): https://papers.miccai.org/miccai-2026-sat/paper/ShapeMI_020.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=BS6fceDaId
BibTex
@InProceedings{MacFra_Segmentation_MICCAISAT2026,
author = { Maccarone, Francesca AND Maggioli, Filippo AND Peruzzo, Denis AND Melzi, Simone},
title = { { Segmentation and keypoints transfer for volumetric medical shapes via functional maps } },
booktitle = {Medical Image Computing and Computer Assisted Intervention -- MICCAI 2026 Workshops and Challenges},
year = {2026},
publisher = {Springer Nature Switzerland},
volume = {LNCS 17259},
month = {pending},
page = {pending}
}
