Paired Image Generation with Diffusion-Guided Diffusion Models

2025-07-20

AI Summary

To address the challenges of low lesion conspicuity and severe scarcity of annotated data in digital breast tomosynthesis (DBT) images—which critically limit segmentation model training—this paper proposes an unconditioned paired diffusion generation method. Leveraging a conditional diffusion framework, we design a novel diffusion guidance network that jointly synthesizes lesion-contaminated DBT images and their corresponding pixel-level segmentation masks within a single denoising step. Crucially, our approach achieves synchronized image-mask generation without requiring paired real data or auxiliary conditional inputs—a first in medical image synthesis. By integrating noise prediction with feature-guided masking, it significantly improves generation fidelity for subtle lesions and mask accuracy. Experiments on DBT slices demonstrate that the synthesized data substantially boost downstream segmentation performance. This work establishes a scalable, annotation-efficient data augmentation paradigm for medical image segmentation under extreme label scarcity.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Deep Generative Models & Autoencoders
• Natural Language Processing: Generation

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Web Mining and Content Analysis: Web data generation and simulation

Abstract

The segmentation of mass lesions in digital breast tomosynthesis (DBT) images is very significant for the early screening of breast cancer. However, the high-density breast tissue often leads to high concealment of the mass lesions, which makes manual annotation difficult and time-consuming. As a result, there is a lack of annotated data for model training. Diffusion models are commonly used for data augmentation, but the existing methods face two challenges. First, due to the high concealment of lesions, it is difficult for the model to learn the features of the lesion area. This leads to the low generation quality of the lesion areas, thus limiting the quality of the generated images. Second, existing methods can only generate images and cannot generate corresponding annotations, which restricts the usability of the generated images in supervised training. In this work, we propose a paired image generation method. The method does not require external conditions and can achieve the generation of paired images by training an extra diffusion guider for the conditional diffusion model. During the experimental phase, we generated paired DBT slices and mass lesion masks. Then, we incorporated them into the supervised training process of the mass lesion segmentation task. The experimental results show that our method can improve the generation quality without external conditions. Moreover, it contributes to alleviating the shortage of annotated data, thus enhancing the performance of downstream tasks.

Problem

• Improving mass lesion segmentation in DBT images for breast cancer screening
• Enhancing paired image generation quality without external conditions
• Addressing annotated data shortage for supervised training in lesion segmentation

Innovation

• Paired image generation without external conditions
• Diffusion guider enhances conditional diffusion model
• Generates both images and corresponding annotations