Few-Shot Generation of Brain Tumors for Secure and Fair Data Sharing

2025-03-31

AI Summary

To address privacy leakage and data scarcity in multi-center brain tumor data sharing, this paper proposes a Decentralized Few-shot Generation Model (DFGM). DFGM avoids uploading raw data or model parameters; instead, it locally synthesizes high-fidelity, controllable tumor images by disentangling private tumor foregrounds from publicly available healthy background images. It integrates diffusion modeling, foreground-background disentanglement, decentralized training, and healthy-image guidance—establishing the first privacy-preserving, few-shot medical image synthesis paradigm. Evaluated on UNet-based segmentation, DFGM-enhanced training improves Dice score by 3.9% and reduces cross-center performance variance by 4.6%, significantly outperforming existing baselines. The method ensures strong data privacy, promotes fairness across institutions, and maintains precise control over synthetic image attributes—including anatomical plausibility and lesion morphology—without compromising security or utility.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Privacy
• Natural Language Processing: Code Generation / Program Synthesis from Natural Language

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Security and Privacy: Data transparency and provenance
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Leveraging multi-center data for medical analytics presents challenges due to privacy concerns and data heterogeneity. While distributed approaches such as federated learning has gained traction, they remain vulnerable to privacy breaches, particularly in sensitive domains like medical imaging. Generative models, such as diffusion models, enhance privacy by synthesizing realistic data. However, they are prone to memorization, especially when trained on small datasets. This study proposes a decentralized few-shot generative model (DFGM) to synthesize brain tumor images while fully preserving privacy. DFGM harmonizes private tumor data with publicly shareable healthy images from multiple medical centers, constructing a new dataset by blending tumor foregrounds with healthy backgrounds. This approach ensures stringent privacy protection and enables controllable, high-quality synthesis by preserving both the healthy backgrounds and tumor foregrounds. We assess DFGM's effectiveness in brain tumor segmentation using a UNet, achieving Dice score improvements of 3.9% for data augmentation and 4.6% for fairness on a separate dataset.

Problem

• Privacy-preserving synthesis of brain tumor images
• Few-shot generation to avoid data memorization
• Harmonizing multi-center data for fair medical analytics

Innovation

• Decentralized few-shot generative model for privacy
• Blends tumor foregrounds with healthy backgrounds
• Enhances segmentation with data augmentation fairness