Tumor Synthesis conditioned on Radiomics

2025-09-28

AI Summary

Medical 3D imaging data are scarce due to privacy constraints, and existing generative models suffer from limited diversity and insufficient biological interpretability. Method: We propose the first radiomics-conditioned controllable tumor synthesis framework for 3D medical images. It jointly employs a GAN to generate anatomically consistent tumor masks and a conditional diffusion model to synthesize high-fidelity tumor textures adhering to quantitative radiomic features—including size, shape, and texture—enabling precise tumor localization, editing, and removal. Contribution/Results: This work pioneers the use of quantitative radiomic features as explicit generative priors, enabling biologically meaningful and highly controllable 3D tumor synthesis. Validated across multi-center datasets of renal, pulmonary, breast, and brain cancers, the synthesized images exhibit high realism and strong expert consensus, significantly improving downstream segmentation and diagnostic model performance.

Technology Category

• Computer Vision: Generative Adversarial Networks (GANs) for Vision
• Machine Learning: Deep Generative Models & Autoencoders
• Natural Language Processing: Generation

Application Category

• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Web Mining and Content Analysis: Web data generation and simulation
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Due to privacy concerns, obtaining large datasets is challenging in medical image analysis, especially with 3D modalities like Computed Tomography (CT) and Magnetic Resonance Imaging (MRI). Existing generative models, developed to address this issue, often face limitations in output diversity and thus cannot accurately represent 3D medical images. We propose a tumor-generation model that utilizes radiomics features as generative conditions. Radiomics features are high-dimensional handcrafted semantic features that are biologically well-grounded and thus are good candidates for conditioning. Our model employs a GAN-based model to generate tumor masks and a diffusion-based approach to generate tumor texture conditioned on radiomics features. Our method allows the user to generate tumor images according to user-specified radiomics features such as size, shape, and texture at an arbitrary location. This enables the physicians to easily visualize tumor images to better understand tumors according to changing radiomics features. Our approach allows for the removal, manipulation, and repositioning of tumors, generating various tumor types in different scenarios. The model has been tested on tumors in four different organs (kidney, lung, breast, and brain) across CT and MRI. The synthesized images are shown to effectively aid in training for downstream tasks and their authenticity was also evaluated through expert evaluations. Our method has potential usage in treatment planning with diverse synthesized tumors.

Problem

• Generating diverse 3D medical tumor images with privacy constraints
• Overcoming limited output diversity in existing generative models
• Using radiomics features to conditionally synthesize customizable tumor images

Innovation

• Generates tumor masks using GAN-based model
• Creates tumor texture via diffusion-based approach
• Conditions synthesis on radiomics features like size