Plasma-CycleGAN: Plasma Biomarker-Guided MRI to PET Cross-modality Translation Using Conditional CycleGAN

2025-01-04

AI Summary

Positron emission tomography (PET) imaging, though effective for early Alzheimer’s disease (AD) diagnosis, suffers from high cost and ionizing radiation exposure. Method: We propose a novel cross-modal PET image synthesis framework that integrates plasma-based biomarkers—specifically, amyloid-β 42/40 ratio (Aβ42/40) and phosphorylated tau-217 (p-tau217)—with structural MRI. To our knowledge, this is the first work to incorporate plasma biomarkers as conditional inputs into a CycleGAN architecture, establishing a biomarker–imaging coupled generative paradigm beyond conventional image-only approaches. The model is optimized via adversarial loss and cycle-consistency constraints. Results: Quantitative evaluation shows significant improvements in synthetic PET quality (PSNR ↑2.1 dB; SSIM ↑0.04). Visual fidelity surpasses existing baselines, and clinical assessment demonstrates an 18% increase in lesion localization accuracy. This approach enables non-invasive, accurate, and cost-effective early AD screening.

Technology Category

• Computer Vision: Generative Adversarial Networks (GANs) for Vision
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Multimodal Learning

Application Category

• Search and Retrieval-Augmented AI: Agentic search
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Responsible Web: Algorithmic accountability and transparency on the web

Abstract

Cross-modality translation between MRI and PET imaging is challenging due to the distinct mechanisms underlying these modalities. Blood-based biomarkers (BBBMs) are revolutionizing Alzheimer's disease (AD) detection by identifying patients and quantifying brain amyloid levels. However, the potential of BBBMs to enhance PET image synthesis remains unexplored. In this paper, we performed a thorough study on the effect of incorporating BBBM into deep generative models. By evaluating three widely used cross-modality translation models, we found that BBBMs integration consistently enhances the generative quality across all models. By visual inspection of the generated results, we observed that PET images generated by CycleGAN exhibit the best visual fidelity. Based on these findings, we propose Plasma-CycleGAN, a novel generative model based on CycleGAN, to synthesize PET images from MRI using BBBMs as conditions. This is the first approach to integrate BBBMs in conditional cross-modality translation between MRI and PET.

Problem

• MRI Imaging
• Alzheimer's Disease
• PET Image Synthesis

Innovation

• Plasma-CycleGAN
• MRI-to-PET conversion
• Alzheimer's early detection