Cascaded 3D Diffusion Models for Whole-body 3D 18-F FDG PET/CT synthesis from Demographics

2025-05-28

AI Summary

To address the urgent need for high-fidelity digital twins in oncology imaging, virtual clinical trials, and AI-driven data augmentation, this work proposes the first end-to-end diffusion-based cascade model for generating whole-body 3D PET/CT volumes conditioned on demographic variables (age, sex, BMI). The method comprises two stages: (1) a coarse-stage diffusion model generating low-resolution PET/CT volumes, followed by (2) a residual super-resolution diffusion module that refines anatomical detail and quantitative SUV distribution. Unlike conventional deterministic phantoms, our approach enables population-statistics–driven, subject-specific image synthesis. Quantitative evaluation demonstrates strong agreement with real clinical data: organ volumes and SUV histogram distributions exhibit high fidelity; subgroup metabolic deviations are ≤5%, substantially outperforming existing generative methods. This is the first study to validate synthetic PET/CT images at clinical-grade fidelity across both anatomical and functional dimensions.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Deep Generative Models & Autoencoders
• Natural Language Processing: Code Generation / Program Synthesis from Natural Language

Application Category

• Web Mining and Content Analysis: Web data generation and simulation
• User Modeling, Personalization and Recommendation: Accountability, Transparency, and Ethics for personalization
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI

Abstract

We propose a cascaded 3D diffusion model framework to synthesize high-fidelity 3D PET/CT volumes directly from demographic variables, addressing the growing need for realistic digital twins in oncologic imaging, virtual trials, and AI-driven data augmentation. Unlike deterministic phantoms, which rely on predefined anatomical and metabolic templates, our method employs a two-stage generative process. An initial score-based diffusion model synthesizes low-resolution PET/CT volumes from demographic variables alone, providing global anatomical structures and approximate metabolic activity. This is followed by a super-resolution residual diffusion model that refines spatial resolution. Our framework was trained on 18-F FDG PET/CT scans from the AutoPET dataset and evaluated using organ-wise volume and standardized uptake value (SUV) distributions, comparing synthetic and real data between demographic subgroups. The organ-wise comparison demonstrated strong concordance between synthetic and real images. In particular, most deviations in metabolic uptake values remained within 3-5% of the ground truth in subgroup analysis. These findings highlight the potential of cascaded 3D diffusion models to generate anatomically and metabolically accurate PET/CT images, offering a robust alternative to traditional phantoms and enabling scalable, population-informed synthetic imaging for clinical and research applications.

Problem

• Synthesize high-fidelity 3D PET/CT volumes from demographics
• Address need for realistic digital twins in medical imaging
• Generate anatomically and metabolically accurate synthetic PET/CT images

Innovation

• Cascaded 3D diffusion models for PET/CT synthesis
• Two-stage generative process enhances resolution
• Demographics-driven synthetic imaging with high accuracy