Search papers, labs, and topics across Lattice.
This study investigates the impact of voxel spacing on volumetric medical image segmentation, addressing a gap in the systematic exploration of image resampling's effects on segmentation outcomes. The authors introduce Consispace, a semantic-aware resampling framework that maintains consistent voxel spacing while ensuring anatomical and semantic fidelity through an ODE-based anatomical constraint and feature reweighting. Experimental results across various datasets reveal that Consispace significantly enhances reconstruction quality and downstream segmentation performance compared to traditional methods.
Achieving consistent voxel spacing can dramatically improve medical image segmentation quality, leading to better diagnostic and surgical outcomes.
Volumetric medical image segmentation is essential for both preoperative diagnosis and intraoperative guidance. While recent years have witnessed rapid progress in segmentation architectures, comparatively little attention is paid to the physical voxel spacing of anatomical data. Indeed, volumetric image resampling is a ubiquitous preprocessing step before segmentation, yet its interaction with downstream segmentation has not been systematically exploited. In this work, we study the correlation between image resampling and segmentation, and propose Consispace, a semantic-aware resampling framework that achieves consistent voxel spacing in the axial direction while preserving anatomical and semantic consistency. Consispace introduces an ODE-based anatomical constraint to model inter-slice dynamics with a continuous interpolator, enabling faithful reconstruction under complex anatomical transitions beyond discrete interpolation. To further couple resampling with segmentation objectives, we leverage dense features from a pretrained vision model to build intra-slice semantic correlation maps and inject class-wise semantic consistency via feature reweighting during resampling. Both intra-slice and inter-slice constraints are integrated into an implicit neural network, supporting arbitrary-scale resampling. Extensive experiments on multiple datasets demonstrate that Consispace achieves superior reconstruction quality and perceptual fidelity, produces smoother inter-slice anatomy, and improves downstream segmentation performance when used as a preprocessing step.
