Controllable Histopathology Image Synthesis with Training-free Structural Initialization and Textural Modulation

2026-06-26

AI Summary

This work addresses the challenge of scarce and costly annotated data in histopathological image analysis, where existing synthesis methods still rely on limited annotations to impose structural constraints. The authors propose CHIS, a novel framework that achieves, for the first time, fully annotation-free controllable generation of pathological images. By leveraging phase fusion in the frequency domain for structural initialization and multi-scale wavelet-based texture modulation, CHIS guides a pre-trained diffusion model—without any additional training—to generate images that precisely align with a given structural mask while preserving realistic tissue appearance. Extensive experiments demonstrate that the proposed method substantially outperforms current approaches, with synthesized data showing high fidelity and significantly enhancing performance in downstream segmentation tasks, thereby validating its practical utility.

Abstract

Deep learning has demonstrated remarkable success in high-throughput histopathology image analysis. However, the performance of learning-based models critically depends on the quality and size of annotations by expert pathologists, which is a resource-intensive and time-consuming process. To address the limitations of data scarcity and annotation burden, several methods have been proposed to synthesize paired histopathology data. Nevertheless, these frameworks typically still require annotation data, albeit in reduced quantities, to impose structural constraints during training. In this work, we present CHIS, a plug-in framework that guides the sampling trajectory of a pretrained diffusion model through two key stages: structural initialization at the start and textural modulation during generation. The initial noise state is refined by fusing the phase information from a prior mask with the amplitude of Gaussian noise in the frequency domain, yielding a structurally informed starting point. During the reverse diffusion process, we adaptively modulate both coarse-grained and fine-grained textures at different wavelet decomposition levels. This enables a diffusion model pretrained solely on unlabeled images to generate outputs that align with prior structural masks while preserving the reference tissue style. We conducted extensive experiments demonstrating the superiority of CHIS in generation fidelity and its substantial benefits for downstream segmentation tasks. Code is available at https://github.com/IBIL-Code/CHIS.

Problem

• histopathology image synthesis
• data scarcity
• annotation burden
• structural control
• training-free generation

Innovation

• training-free
• structural initialization
• textural modulation
• diffusion model
• histopathology image synthesis