The paper introduces a geometry‑guided sampling operator that directs feature sampling rather than altering convolution kernels in 3D encoder‑decoder networks. By predicting local orientations and bounded step sizes, the operator samples symmetrically around each voxel, generating compact geometric and boundary cues that improve fine‑structure segmentation. Replacing stride‑1 and stride‑2 operations in a 3D U‑Net yields consistent gains on BraTS, MSD Hepatic Vessel, and TDSC‑ABUS datasets, with better boundary metrics and fewer parameters, and the operator can be integrated into other backbones without architectural changes.
arXiv:2610.20509v1 Announce Type: cross Abstract: Generative super-resolution (SR) can recover sharp detail in computed tomography (CT) images from severely downsampled observations, but the missing...
arXiv:2606. 13188v1 Announce Type: cross Abstract: Building patient-specific cardiac models sits at the heart of precision cardiology, yet getting those models into clinical use keeps running into the same wall: mesh generation is slow, messy, and frustrating.
arXiv:2605.30972v2 Announce Type: replace Abstract: Accurate 3D medical image segmentation requires both fine spatial detail and long-range volumetric context. Although Mamba provides efficient long-...
arXiv:2610.06938v1 Announce Type: new Abstract: Medical image segmentation remains fragmented along two axes: segmentation paradigms and data dimensionality. Existing methods are typically developed...
arXiv:2608. 19965v1 Announce Type: cross Abstract: Segmentation of curvilinear anatomical structures in 3D medical images remains challenging due to complex topology, severe class imbalance, weak contrast, and large variations in structure morphology.
DALE-CT introduces depth‑aware 2D slice encoders that learn an anatomical world model of chest CT scans without 3D or positional supervision. By sampling self‑supervised views across a physical $z$‑axis slab, the encoder captures how anatomy changes between neighboring slices, enabling it to recover slice ordering and distinguish slices by anatomy alone. The model, trained on a large 287k‑scan corpus, achieves state‑of‑the‑art performance on CT‑RATE and is released with full code and evaluation tools.
arXiv:2610. 13915v1 Announce Type: new Abstract: Self-supervised learning on 3D medical images commonly relies on random sub-volume sampling during pretraining.
arXiv:2607. 26829v1 Announce Type: cross Abstract: Many high-performing volumetric segmentation models maintain dense multi-scale feature maps, leading to high activation memory and inference cost.
arXiv:2608. 15537v1 Announce Type: cross Abstract: Accurate boundary delineation remains a persistent challenge in dermoscopic image segmentation because of blurred lesion margins, heterogeneous textures, and complex background artifacts.
Accurate medical image interpolation and anatomical structure segmentation are fundamental for computer-aided diagnosis and treatment planning. Anisotropic medical volumes with sparse through-plane sampling often suffer from structural discontinuity and boundary blur, hindering reliable clinical image analysis.
ContiLNN augments 2‑D medical image restoration backbones with bidirectional closed‑form continuous‑time (Bi‑CfC) modules to model cross‑slice relationships while preserving in‑plane feature extraction. The method uses slice‑index intervals to modulate gates based on local features and hidden states, enabling propagation that adapts to slice‑sampling variations without numerical ODE integration. Across CT denoising, MRI super‑resolution, and PET restoration tasks, ContiLNN outperforms Restore‑RWKV and Bi‑GRU baselines in PSNR, RMSE, latency, and GPU memory usage, demonstrating improved fidelity and data efficiency.
The growing number of medical vision foundation models highlights the need for effective model selection. However, mainstream selection methods rely on exhaustive fine-tuning, which is computationally expensive.