The paper introduces a novel 3‑D reconstruction framework for bilateral coronary artery trees from cardiac CT angiography, addressing limitations such as voxel quantization, territory ambiguity, and ostium drift. It employs a canonical SO(3) multi‑scale representation, a Riemannian arborescence solver, and a sub‑voxel displacement head, training 327,385 active parameters in a single 30‑epoch AdamW loop. On unseen datasets, the method achieves sub‑millimetre accuracy in centerline distance and ostial localisation, outperforming prior approaches.
By Zeyu Jia, Xin Ming
arXiv:2608. 12274v1 Announce Type: cross Abstract: Background: Accurate segmentation of the Left Anterior Descending (LAD) artery in 3D free-breathing, non-contrast CT is critical for cardiac dose sparing in thoracic radiotherapy.
By Rafi Ibn Sultan, Chengyin Li, Yiannos Demetriou, Ahmed I. Ghanem, Joshua P. Kim, Justine Cunningham, Hassan Bagher-Ebadian, Dongxiao Zhu, Kundan S. Thind
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.
By Abhishek H S, Akash Ganamukhi, Abhimanyu Suresh, Aditya G Hiremath, Prasad B Honnavalli, Adithya Balasubramanyam
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.
By Sizhe Wang, Himashi Peiris, Zhaolin Chen
arXiv:2611.05693v1 Announce Type: new
Abstract: Sli2Vol propagates a binary mask through affinities between adjacent slices, but an error can persist because each prediction becomes the input to the...
By Tushar Kataria, Md Rakibul Haque, Shireen Y. Elhabian
arXiv:2607. 22718v1 Announce Type: cross Abstract: We propose parameter-efficient SSM-based U-Net architectures for 3D medical image segmentation.
By Mohammad Arafat Hussain, Ellen Grant, Yangming Ou