analiza aiqbench
MAGEFormer addresses the metric mismatch in volumetric medical imaging by embedding physical constraints directly into Vision Transformers, bypassing the need for isotropic resampling that degrades accuracy. By integrating Metric-Adaptive Spatial Embedding and Geometry-Constrained Attention, this approach achieves superior boundary precision on standard abdominal CT benchmarks compared to conventional CNN and Transformer baselines. This work demonstrates that geometry-aware representation learning is critical for handling anisotropic data in clinical segmentation tasks.
MAGEFormer: Learning Metric-Consistent Representations for Anisotropic CT Segmentation
arXiv:2610.04036v1 Announce Type: cross Abstract: Vision Transformers (ViTs) have shown strong performance in volumetric segmentation, but their effectiveness on clinical CT is limited by an isotropic Euclidean lattice assumption. This conflicts with anisotropic CT acquisition, leading to two key issues: (1) a metric mismatch between voxel indices and physical anatomy, and (2) accuracy degradation from isotropic resampling. To address this, we propose MAGEFormer, a geometry-calibrated framework that embeds physical metric constraints directly into representation learning. Our method introduces Metric-Adaptive Spatial Embedding (MASE) to calibrate positional frequencies using voxel spacing, Geometry-Constrained Attention (GCA) to suppress physically implausible feature correlations, and Geometric View Voting (GVV) to reduce discretization bias during inference. We evaluate MAGEFormer on two multi-organ abdominal CT benchmarks, BTCV and FLARE 22, under a unified protocol against strong CNN and Transformer-based baselines. MAGEFormer achieves the strongest boundary accuracy among the compared methods, with 10.58 mm HD95 on BTCV and 3.40 mm HD95 on FLARE 22, and shows consistent gains in Dice under the same protocol. These results show that geometry-aware internal calibration is more effective than relying on conventional isotropic preprocessing alone for anisotropic CT segmentation.
Abstract:Vision Transformers (ViTs) have shown strong performance in volumetric segmentation, but their effectiveness on clinical CT is limited by an isotropic Euclidean lattice assumption. This conflicts with anisotropic CT acquisition, leading to two key issues: (1) a metric mismatch between voxel indices and physical anatomy, and (2) accuracy degradation from isotropic resampling. To address this, we propose MAGEFormer, a geometry-calibrated framework that embeds physical metric constraints directly into representation learning. Our method introduces Metric-Adaptive Spatial Embedding (MASE) to calibrate positional frequencies using voxel spacing, Geometry-Constrained Attention (GCA) to suppress physically implausible feature correlations, and Geometric View Voting (GVV) to reduce discretization bias during inference. We evaluate MAGEFormer on two multi-organ abdominal CT benchmarks, BTCV and FLARE 22, under a unified protocol against strong CNN and Transformer-based baselines. MAGEFormer achieves the strongest boundary accuracy among the compared methods, with 10.58 mm HD95 on BTCV and 3.40 mm HD95 on FLARE 22, and shows consistent gains in Dice under the same protocol. These results show that geometry-aware internal calibration is more effective than relying on conventional isotropic preprocessing alone for anisotropic CT segmentation.
Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
arXiv:2610.04036 [cs.CV]
(or arXiv:2610.04036v1 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2610.04036 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
Medical Image Computing and Computer Assisted Intervention - MICCAI 2026, Part VII, LNCS 16884, pp. 327-337 (2027)
https://doi.org/10.1007/978-3-032-38085-2_31 Focus to learn more DOI(s) linking to related resources
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 308 cuvinte
Subiecte conexe
Contrastive Metric Learning for Point Cloud Segmentation in Highly Granular Detectors
arXiv · 15 oct. 2026
Modernizing and scaling support operations with generative AI on AWS
AWS · 2 sept. 2026
A Quantitative Evaluation Framework for Temporal Explainability in Echocardiographic Video Segmentation
arXiv · 9 sept. 2026
SAM-V: Geometry-Aware Segment Anything for Multi-View Instance Segmentation
arXiv · 23 sept. 2026