Structure-Guided Histopathology Synthesis via Dual-LoRA Diffusion
This work addresses the problem of realistic and structure-consistent histopathology image synthesis for tissue restoration, data augmentation, and tumor microenvironment modeling, offering a unified solution for researchers and practitioners in medical imaging.
This paper proposes Dual-LoRA Controllable Diffusion, a unified centroid-guided diffusion framework for histopathology image synthesis. It jointly supports local structure completion and global structure synthesis, achieving an LPIPS improvement from 0.1797 to 0.1524 for local completion and an FID improvement from 225.15 to 76.04 for global synthesis compared to state-of-the-art baselines.
Histopathology image synthesis plays an important role in tissue restoration, data augmentation, and modeling of tumor microenvironments. However, existing generative methods typically address restoration and generation as separate tasks, although both share the same objective of structure-consistent tissue synthesis under varying degrees of missingness, and often rely on weak or inconsistent structural priors that limit realistic cellular organization. We propose Dual-LoRA Controllable Diffusion, a unified centroid-guided diffusion framework that jointly supports Local Structure Completion and Global Structure Synthesis within a single model. Multi-class nuclei centroids serve as lightweight and annotation-efficient spatial priors, providing biologically meaningful guidance under both partial and complete image absence. Two task-specific LoRA adapters specialize the shared backbone for local and global objectives without retraining separate diffusion models. Extensive experiments demonstrate consistent improvements over state-of-the-art GAN and diffusion baselines across restoration and synthesis tasks. For local completion, LPIPS computed within the masked region improves from 0.1797 (HARP) to 0.1524, and for global synthesis, FID improves from 225.15 (CoSys) to 76.04, indicating improved structural fidelity and realism. Our approach achieves more faithful structural recovery in masked regions and substantially improved realism and morphology consistency in full synthesis, supporting scalable pan-cancer histopathology modeling.