High-Fidelity Medical Shape Generation via Skeletal Latent Diffusion

Guoqing Zhang; Jingyun Yang; Siqi Chen; Anping Zhang; Yang Li

cs.CVMar 8, 2026

High-Fidelity Medical Shape Generation via Skeletal Latent Diffusion

Authors: Guoqing Zhang, Jingyun Yang, Siqi Chen, Anping Zhang, Yang Li

Organizations: Tsinghua Shenzhen International Graduate School, Tsinghua University, Shenzhen, China · Pengcheng Laboratory, Shenzhen, China · School of AI, Chinese University of Hong Kong (Shenzhen)

Abstract

Anatomy shape modeling is a fundamental problem in medical data analysis. However, the geometric complexity and topological variability of anatomical structures pose significant challenges to accurate anatomical shape generation. In this work, we propose a skeletal latent diffusion framework that explicitly incorporates structural priors for efficient and high-fidelity medical shape generation. We introduce a shape auto-encoder in which the encoder captures global geometric information through a differentiable skeletonization module and aggregates local surface features into shape latents, while the decoder predicts the corresponding implicit fields over sparsely sampled coordinates. New shapes are generated via a latent-space diffusion model, followed by neural implicit decoding and mesh extraction. To address the limited availability of medical shape data, we construct a large-scale dataset, MedSDF, comprising surface point clouds and corresponding signed distance fields across multiple anatomical categories. Extensive experiments on MedSDF and vessel datasets demonstrate that the proposed method achieves superior reconstruction and generation quality while maintaining a higher computational efficiency compared with existing approaches. Code is available at: https://github.com/wlsdzyzl/meshage.

Figures & tables

Fig. 1: A semantic overview of the proposed architecture. We first train a VAE where the encoder maps the input surface and online-computed skeleton to a latent point set, which serves as a shape signature for the decoder to predict SDF values at queried spatial coordinates. A diffusion model is then trained in the latent feature space. For shape synthesis, new latents are sampled via the reverse diffusion process and decoded into neural implicit fields, which are subsequently converted into 3D shapes.

Fig. 2: Left: differentiable skeletonization. Right: skeleton-guided coordinate sampling (right); only $10\%$ voxels (green grids) near the skeletal points (blue dots) are sampled for SDF calculation during inference.

Fig. 3: Qualitative comparison of shape reconstruction on MedSDF dataset. From top to bottom, the rows correspond to samples of the bladder, brain, colon, right coronary artery (RCA), left coronary artery (LCA), duodenum, gallbladder, liver, pancreas, spleen, stomach, and trachea. GeM3D (S) and Ours (S) denotes the skeletal points computed by GeM3D and our method, respectively.

Fig. 4: Qualitative comparison of shape generation on MedSDF dataset. From top to bottom, the rows correspond to samples of the bladder, brain, colon, right coronary artery (RCA), left coronary artery (LCA), duodenum, gallbladder, liver, pancreas, spleen, stomach, and trachea. For each category and method, we generate 500 samples and display the two nearest neighbors of the reference point cloud.

Method	CD ( $\downarrow$ )	EMD ( $\downarrow$ )	HD ( $\downarrow$ )	F1 ( $\%,\uparrow$ )
PointNet++	3.811	1.849	2.045	85.97
DGCNN	3.681	1.793	2.104	87.29
PCT	5.577	3.875	3.029	67.98
Point-SSM	3.876	2.012	2.336	85.30
Diff-PCD	6.001	1.278	3.023	61.47
GeM3D	2.672	0.643	1.260	97.86

TABLE I: Shape reconstruction evaluation on the MedSDF dataset. CD is multiplied by

10^{2}

, EMD is multiplied by

10

, and HD is multiplied by

10

. The best results are denoted in Bold .

Metric	Diff-PCD	EDM-PCD	PVD	GeM3D	Ours
FID ( $\downarrow$ )	95.83	51.50	66.35	133.7	35.99
KID ( $\downarrow$ )	10.99	3.544	5.057	15.69	2.204
MMD-CD ( $\downarrow$ )	6.923	8.044	8.411	9.537	7.720
MMD-EMD ( $\downarrow$ )	0.941	1.002	1.049	1.095	0.930
COV-CD ( $\uparrow$ )	70.60	58.06	48.16	38.11	72.87
COV-EMD ( $\uparrow$ )	67.87	57.05	48.02	39.16	72.00

TABLE II: Category-conditioned shape generation evaluation on the MedSDF dataset. KID is multiplied by

10

, MMD-CD is multiplied by

10^{2}

, and MMD-EMD is multiplied by

10

. The best results are denoted in Bold . The second-best score is underlined .

Dataset	Method	Reconstruction				Generation
		CD ( $\downarrow$ )	EMD ( $\downarrow$ )	HD ( $\downarrow$ )	F1 ( $\%,\uparrow$ )	FID ( $\downarrow$ )	KID ( $\downarrow$ )	MMD ( $\downarrow$ )		COV ( $\%,\uparrow$ )		1-NNA ( $\%,\downarrow$ )
		CD ( $\downarrow$ )	EMD ( $\downarrow$ )	HD ( $\downarrow$ )	F1 ( $\%,\uparrow$ )	FID ( $\downarrow$ )	KID ( $\downarrow$ )	CD	EMD	CD	EMD	CD	EMD
CoW	Diff-Vessel	4.907	1.041	2.502	71.79	23.08	1.422	4.978	0.733	67.27	52.72	91.17	91.35
CoW	Ours	1.796	0.321	0.840	99.78	18.28	0.725	5.197	0.749	81.82	80.00	88.64	89.36
ImageCAS	Diff-Vessel	5.810	1.364	3.485	65.42	34.26	1.804	8.471	1.130	59.50	54.50	84.14	82.14
ImageCAS	Ours	1.314	0.272	0.428	99.98	24.18	1.432	7.377	0.975	76.00	66.50	64.85	64.28

TABLE III: Shape reconstruction and generation evaluation on two vessel datasets: CoW and ImageCAS. The best results are denoted in Bold .

Fig. 5: Qualitative comparison of tubular shape reconstruction and generation. The top two rows correspond to reconstruction results of input point clouds from CoW and ImageCAS datasets, while the bottom two rows display the neighbors of input point clouds among the generated shapes.

Dataset	Module			Reconstruction				Generation
	DS	SC	LA	CD ( $\downarrow$ )	EMD ( $\downarrow$ )	HD ( $\downarrow$ )	F1 ( $\%,\uparrow$ )	FID ( $\downarrow$ )	KID ( $\downarrow$ )	MMD ( $\downarrow$ )		COV ( $\%,\uparrow$ )		1-NNA ( $\%,\downarrow$ )
	DS	SC	LA	CD ( $\downarrow$ )	EMD ( $\downarrow$ )	HD ( $\downarrow$ )	F1 ( $\%,\uparrow$ )	FID ( $\downarrow$ )	KID ( $\downarrow$ )	CD	EMD	CD	EMD	CD	EMD
CoW	✗	✗	✗	4.102	0.822	1.847	80.39	34.77	4.106	5.249	0.741	81.82	80.00	92.61	91.89
	✔	✗	✗	2.348	0.481	1.293	98.13	23.57	1.844	4.911	0.725	76.36	72.72	91.53	91.35
	✔	✔	✗	2.277	0.487	1.222	98.23	22.63	1.472	4.988	0.734	76.36	74.54	91.89	90.99
	✔	✔	✔	1.796	0.321	0.839	99.78	18.28	0.725	5.197	0.749	81.82	80.00	88.64	89.36

TABLE IV: Ablation study on the CoW, ImageCAS, and MedSDF dataset. The best results are denoted in Bold . Recommended settings are marked with grey.

3D Shape Generation

3D shape generation creates and edits 3D objects from learned geometric priors and text, images, sketches, or structural constraints.

See all 106 papers

MorphoFlow: Sparse-Supervised Generative Shape Modeling with Adaptive Latent Relevance
Apr 13, 2026Mokshagna Sai Teja Karanam, Tushar Kataria, Shireen ElhabianStatistical Shape Modeling 3D Shape Generation
Kahlert School of Computing, Scientific Computing and Imaging Institute, University of Utah, Utah, USA
Sketch2CT: Multimodal Diffusion for Structure-Aware 3D Medical Volume Generation
Mar 23, 2026Delin An, Chaoli WangMedical Image Synthesis 3D Medical Imaging
University of Notre Dame
VoxStruct3D: Structure-Leading Flow Matching for Voxel-Space 3D MRI Synthesis
Aug 5, 2026Fang Li, Yang Gao, Shihao Zou +5Flow Matching 3D Medical Imaging
State Key Laboratory of Virtual Reality Technology and Systems, Beihang University · Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences · Faculty of Computer Science and Control Engineering, Shenzhen University of Advanced Technology

Apr 13, 2026J/K move · Enter open · S save

MorphoFlow: Sparse-Supervised Generative Shape Modeling with Adaptive Latent Relevance

Mokshagna Sai Teja Karanam, Tushar Kataria, Shireen Elhabian

Kahlert School of Computing, Scientific Computing and Imaging Institute, University of Utah, Utah, USA

Statistical Shape Modeling 3D Shape Generation Flow-Based Generative Modeling

Open paper AskPDF arXiv

Statistical shape modeling (SSM) is central to population level analysis of anatomical variability, yet most existing approaches rely on densely annotated segmentations and fixed latent representations. These requirements limit scalability and reduce flexibility when modeling complex anatomical variation. We introduce MorphoFlow, a sparse supervised generative shape modeling framework that learns compact probabilistic shape representations directly from sparse surface annotations. MorphoFlow integrates neural implicit shape representations with an autodecoder formulation and autoregressive normalizing flows to learn an expressive probabilistic density over the latent shape space. The neural implicit representation enables resolution-agnostic modeling of 3D anatomy, while the autodecoder formulation supports direct optimization of per-instance latent codes under sparse supervision. The autoregressive flow captures the distribution of latent anatomical variability providing a tractable, likelihood-based generative model of shapes. To promote compact and structured latent representations, we incorporate adaptive latent relevance weighting through sparsity-inducing priors, enabling the model to regulate the contribution of individual latent dimensions according to their relevance to the underlying anatomical variation while preserving generative expressivity. The resulting latent space supports uncertainty quantification and anatomically plausible shape synthesis without manual latent dimensionality tuning. Evaluation on publicly available lumbar vertebrae and femur datasets demonstrates accurate high-resolution reconstruction from sparse inputs and recovery of structured modes of anatomical variation consistent with population level trends.

Abstract

3D Shape Generation

Explore similar work

MorphoFlow: Sparse-Supervised Generative Shape Modeling with Adaptive Latent Relevance

Sketch2CT: Multimodal Diffusion for Structure-Aware 3D Medical Volume Generation

VoxStruct3D: Structure-Leading Flow Matching for Voxel-Space 3D MRI Synthesis