AbdomenGen: Sequential Volume-Conditioned Diffusion Framework for Abdominal Anatomy Generation

Yubraj Bhandari; Lavsen Dahal; Paul Segars; Joseph Y. Lo

cs.CVApr 14, 2026

AbdomenGen: Sequential Volume-Conditioned Diffusion Framework for Abdominal Anatomy Generation

Authors: Yubraj Bhandari, Lavsen Dahal, Paul Segars, Joseph Y. Lo

Organizations: Center for Virtual Imaging Trials, Dept. of Radiology, Duke University School of Medicine · Dept. of Physics, Trinity School of Arts and Sciences, Duke University · Dept. of Electrical and Computer Engineering, Pratt School of Engineering, Duke University

Abstract

Computational phantoms are widely used in medical imaging research, yet current systems to generate controlled, clinically meaningful anatomical variations remain limited. We present AbdomenGen, a sequential volume-conditioned diffusion framework for controllable abdominal anatomy generation. We introduce the Volume Control Scalar (VCS), a standardized residual that decouples organ size from body habitus, enabling interpretable volume modulation. Organ masks are synthesized sequentially, conditioning on the body mask and previously generated structures to preserve global anatomical coherence while supporting independent, multi-organ control. Across 11 abdominal organs, the proposed framework achieves strong geometric fidelity (e.g., liver dice $0.83 \pm 0.05$ ), stable single-organ calibration over $[-3,+3]$ VCS, and disentangled multi-organ modulation. To showcase clinical utility with a hepatomegaly cohort selected from MERLIN, Wasserstein-based VCS selection reduces distributional distance of training data by 73.6% . These results demonstrate calibrated, distribution-aware anatomical generation suitable for controllable abdominal phantom construction and simulation studies.

Figures & tables

Figure 1: Overview of AbdomenGen, sequential organ generation framework based on denoising diffusion U-Net with FiLM conditioning. For generation of organ $o_{i}$ (illustrated left to right for colon), the model $p_{\theta_{i}}$ takes as input the body mask signed distance field (SDF) $B$ , current context SDF $S_{i}^{\mathrm{ctx}}$ , and volume control scalar $v_{o}$ . The scalar is injected through a FiLM adapter to modulate intermediate network features. The generated organ SDF $\hat{S}_{o_{i}}$ is obtained and incorporated into the context to form $S_{i+1}^{\mathrm{ctx}}$ . Organs are generated in a predefined order (bottom), and the accumulated outputs form the final full abdominal anatomy.

(A) Fidelity: Generated vs Reference Masks
	Liver	Spleen	Kidneys	Stomach
Dice $\uparrow$	0.83 $\pm$ 0.05	0.68 $\pm$ 0.13	0.60 $\pm$ 0.23	0.68 $\pm$ 0.15
ASSD (mm) $\downarrow$	5.85 $\pm$ 2.34	6.23 $\pm$ 2.83	6.16 $\pm$ 6.72	6.36 $\pm$ 4.16
HD95 (mm) $\downarrow$	17.58 $\pm$ 9.35	18.61 $\pm$ 9.05	16.05 $\pm$ 22.50	21.26 $\pm$ 17.76
Chamfer (mm) $\downarrow$	11.70 $\pm$ 4.68	12.47 $\pm$ 5.67	12.33 $\pm$ 13.44	12.71 $\pm$ 8.33
(B) Manifold realism: Generated vs Train-Nearest Neighbor

Table 1: (A) Geometric agreement between generated masks and reference mask (Gen vs Ref). (B) Manifold realism: distance from generated masks to their nearest training sample (Gen vs Train-NN). (C) Reference training variability: distance from reference masks to their nearest training sample (Ref vs Train-NN). Values are reported as mean

\pm

std across 200 test cases.

Model	Pairwise Dice $\downarrow$	Pairwise Chamfer(mm) $\uparrow$
MAISI (size parameter=0.9)	$0.803\pm 0.057$	$13.97\pm 4.31$
Ours (Ref Masks VCS)	0.755 $\pm$ 0.096	17.01 $\pm$ 6.82

Table 2: Diversity comparison between AbdomenGen and MAISI. Values for generated liver distributions are reported as mean

\pm

standard deviation.

Figure 2: Qualitative evaluation: Rows correspond to Liver (cyan, anteroposterior view) and Spleen (pink, posteroanterior view). Columns (left to right) show Reference Masks and VCS $=-2$ , 0, and +2 respectively. As VCS varies, target organ volume (below each rendering) exhibits controlled changes while surrounding anatomical structures remain spatially coherent.

Panel of Figure 3 — Figure 3: (a) Mean organ volume change ( $\Delta$ %) relative to VCS = 0 across 200 test cases (mean ± 95% CI) as VCS is swept over [-3, 3]. Volumes are normalized to baseline generation to enable cross-organ comparison. (b) Kernel density estimates of liver volumes for the training cohort (dotted), Merlin hepatomegaly Reference Masks (solid), and generated samples under the selected volume-conditioning setting (dashed). Vertical dashed lines denote cohort means.

Figure 4: Joint liver–spleen volume distributions under independent VCS conditioning (±2). Gray contours denote the empirical training distribution; colored clusters show diverse but controlled generations that sampling these two dimensions.

Medical Image Synthesis

Medical image synthesis creates, translates, reconstructs, and edits clinical images to fill missing data and support analysis.

See all 310 papers

PIVM: Diffusion-Based Prior-Integrated Variation Modeling for Anatomically Precise Abdominal CT Synthesis
Mar 23, 2026Dinglun He, Baoming Zhang, Xu Wang +3Medical Image Synthesis Computed Tomography
Department of Electrical Engineering and Computer Science, University of Missouri, Columbia, MO, USA · Department of Computer Science, The University of Texas at Dallas, Richardson, TX, USA · Department of Radiation Oncology, Washington University in St. Louis, MO, USA +2
MedSyn2: Flexible Control of 3D CT Generation via Text and Semantically-Defined Segmentation Prompts
May 31, 2026Weicheng Dai, Chenyu Wang, Binxu Li +4Diffusion Transformer Conditional Image Generation
Boston University School of Engineering, Boston, MA, USA · Stanford University · University of Pittsburgh Medical Center +1
NV-Generate-Body: 3D Latent Diffusion Model forControllable Anatomy, Tumor, and Demographic Body Mask Synthesis
Oct 16, 2026Can Zhao, Pengfei Guo, Dong Yang +8Medical Image Synthesis Conditional Generation
NVIDIA · National Institutes of Health · University of Arkansas for Medical Sciences

Mar 23, 2026J/K move · Enter open · S save

PIVM: Diffusion-Based Prior-Integrated Variation Modeling for Anatomically Precise Abdominal CT Synthesis

Dinglun He, Baoming Zhang, Xu Wang, Yao Hao+2

Department of Electrical Engineering and Computer Science, University of Missouri, Columbia, MO, USA · Department of Computer Science, The University of Texas at Dallas, Richardson, TX, USA · Department of Radiation Oncology, Washington University in St. Louis, MO, USA+2

Medical Image Synthesis Computed Tomography Diffusion-Based Image Generation

Open paper AskPDF arXiv

Abdominal CT data are limited by high annotation costs and privacy constraints, which hinder the development of robust segmentation and diagnostic models. We present a Prior-Integrated Variation Modeling (PIVM) framework, a diffusion-based method for anatomically accurate CT image synthesis. Instead of generating full images from noise, PIVM predicts voxel-wise intensity variations relative to organ-specific intensity priors derived from segmentation labels. These priors and labels jointly guide the diffusion process, ensuring spatial alignment and realistic organ boundaries. Unlike latent-space diffusion models, our approach operates directly in image space while preserving the full Hounsfield Unit (HU) range, capturing fine anatomical textures without smoothing. Source code is available at https://github.com/BZNR3/PIVM.

Abstract

Medical Image Synthesis

Explore similar work

PIVM: Diffusion-Based Prior-Integrated Variation Modeling for Anatomically Precise Abdominal CT Synthesis

MedSyn2: Flexible Control of 3D CT Generation via Text and Semantically-Defined Segmentation Prompts

NV-Generate-Body: 3D Latent Diffusion Model forControllable Anatomy, Tumor, and Demographic Body Mask Synthesis