Score-based Generative Diffusion Models to Synthesize Full-dose FDG Brain PET from MRI in Epilepsy Patients

2025-06-12

AI Summary

Children with epilepsy undergoing FDG-PET face non-negligible ionizing radiation exposure, necessitating dose reduction without compromising diagnostic accuracy. Method: We propose a novel MRI/ultra-low-dose PET–driven synthesis framework for full-dose PET images using score-based generative diffusion models (SGM-KD and SGM-VP). For the first time, diffusion models are applied to epilepsy-specific PET synthesis. We introduce a metabolism-asymmetry–aware congruence metric (Congruence Index and MAE) and employ a Transformer-U-Net architecture to jointly encode T1-weighted/T2-FLAIR MRI and 1%–dose PET inputs, optimizing SUVR mapping and interhemispheric symmetry modeling. Results: With MRI-only input, SGM-KD achieves SUVR MAE = 0.021 and ICC = 0.94; incorporating 1%–dose PET yields comparable performance across all models. Synthesized images meet clinical diagnostic standards in both qualitative assessment and quantitative metrics, enabling >99% radiation dose reduction in pediatric epilepsy imaging.

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

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

Fluorodeoxyglucose (FDG) PET to evaluate patients with epilepsy is one of the most common applications for simultaneous PET/MRI, given the need to image both brain structure and metabolism, but is suboptimal due to the radiation dose in this young population. Little work has been done synthesizing diagnostic quality PET images from MRI data or MRI data with ultralow-dose PET using advanced generative AI methods, such as diffusion models, with attention to clinical evaluations tailored for the epilepsy population. Here we compared the performance of diffusion- and non-diffusion-based deep learning models for the MRI-to-PET image translation task for epilepsy imaging using simultaneous PET/MRI in 52 subjects (40 train/2 validate/10 hold-out test). We tested three different models: 2 score-based generative diffusion models (SGM-Karras Diffusion [SGM-KD] and SGM-variance preserving [SGM-VP]) and a Transformer-Unet. We report results on standard image processing metrics as well as clinically relevant metrics, including congruency measures (Congruence Index and Congruency Mean Absolute Error) that assess hemispheric metabolic asymmetry, which is a key part of the clinical analysis of these images. The SGM-KD produced the best qualitative and quantitative results when synthesizing PET purely from T1w and T2 FLAIR images with the least mean absolute error in whole-brain specific uptake value ratio (SUVR) and highest intraclass correlation coefficient. When 1% low-dose PET images are included in the inputs, all models improve significantly and are interchangeable for quantitative performance and visual quality. In summary, SGMs hold great potential for pure MRI-to-PET translation, while all 3 model types can synthesize full-dose FDG-PET accurately using MRI and ultralow-dose PET.

Problem

• Synthesize full-dose FDG brain PET from MRI using diffusion models
• Reduce radiation dose in epilepsy patients with generative AI
• Evaluate clinical metrics for epilepsy-specific PET image synthesis

Innovation

• Score-based generative diffusion models for PET synthesis
• MRI-to-PET translation with minimal radiation dose
• Clinical evaluation tailored for epilepsy patients