# 3D-LLDM: Label-Guided 3D Latent Diffusion Model for Improving High-Resolution Synthetic MR Imaging in Hepatic Structure Segmentation

arXiv: https://arxiv.org/abs/2603.23845 · HTML: https://mlsanity.com/paper/2603.23845/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Medical imaging and segmentation, one of 14 directions in Diffusion models.
- Builds on 18 of its 31 references that are in the map.
- First posted on arXiv 25 Mar 2026.

## Placed in

- Diffusion models / [Medical imaging and segmentation](https://mlsanity.com/diffusion/medical-imaging-and-segmentation/)

## Builds on (most cited first, 18 of 18)

- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Adding Conditional Control to Text-to-Image Diffusion Models](https://mlsanity.com/paper/2302.05543/) (2023)
- [Diffusion Models in Vision: A Survey](https://mlsanity.com/paper/2209.04747/) (2022)
- [Image Segmentation Using Deep Learning: A Survey](https://mlsanity.com/paper/2001.05566/) (2020)
- [MONAI: An open-source framework for deep learning in healthcare](https://mlsanity.com/paper/2211.02701/) (2022)
- [Road Extraction by Deep Residual U-Net](https://mlsanity.com/paper/1711.10684/) (2017)
- [Med3D: Transfer Learning for 3D Medical Image Analysis](https://mlsanity.com/paper/1904.00625/) (2019)
- [Diffusion Models for Medical Image Analysis: A Comprehensive Survey](https://mlsanity.com/paper/2211.07804/) (2022)
- [CLIP-Driven Universal Model for Organ Segmentation and Tumor Detection](https://mlsanity.com/paper/2301.00785/) (2023)
- [GenerateCT: Text-Conditional Generation of 3D Chest CT Volumes](https://mlsanity.com/paper/2305.16037/) (2023)
- [Label-Assemble: Leveraging Multiple Datasets with Partial Labels](https://mlsanity.com/paper/2109.12265/) (2021)
- [From Pixel to Cancer: Cellular Automata in Computed Tomography](https://mlsanity.com/paper/2403.06459/) (2024)
- [FreeTumor: Advance Tumor Segmentation via Large-Scale Tumor Synthesis](https://mlsanity.com/paper/2406.01264/) (2024)
- [Synthetic Tumors Make AI Segment Tumors Better](https://mlsanity.com/paper/2210.14845/) (2022)
- [3D MRI Synthesis with Slice-Based Latent Diffusion Models: Improving Tumor Segmentation Tasks in Data-Scarce Regimes](https://mlsanity.com/paper/2406.05421/) (2024)
- [HAGAN: Hybrid Augmented Generative Adversarial Network for Medical Image Synthesis](https://mlsanity.com/paper/2405.04902/) (2024)

## Built on by (most cited first, 2 of 2)

- [Adaptive Routing for Efficient Diffusion Transformer-Based PNI Prediction](https://mlsanity.com/paper/2607.11533/) (2026)
- [SCINTILLA-SNN: A Spiking Multi-Scale Selective Aggregation Network for Perineural Invasion Prediction](https://mlsanity.com/paper/2609.11237/) (2026)

## Abstract (arXiv)

Deep learning and generative models are advancing rapidly, with synthetic data increasingly being integrated into training pipelines for downstream analysis tasks. However, in medical imaging, their adoption remains constrained by the scarcity of reliable annotated datasets. To address this limitation, we propose 3D-LLDM, a label-guided 3D latent diffusion model that generates high-quality synthetic magnetic resonance (MR) volumes with corresponding anatomical segmentation masks. Our approach uses hepatobiliary phase MR images enhanced with the Gd-EOB-DTPA contrast agent to derive structural masks for the liver, portal vein, hepatic vein, and hepatocellular carcinoma, which then guide volumetric synthesis through a ControlNet-based architecture. Trained on 720 real clinical hepatobiliary phase MR scans from Samsung Medical Center, 3D-LLDM achieves a Fr'echet Inception Distance (FID) of 28.31, improving over GANs by 70.9% and over state-of-the-art diffusion baselines by 26.7%. When used for data augmentation, the synthetic volumes improve hepatocellular carcinoma segmentation by up to 11.153% Dice score across five CNN architectures.