MCR-VQGAN: A Scalable and Cost-Effective Tau PET Synthesis Approach for Alzheimer's Disease Imaging

2025-12-17

AI Summary

To address the clinical limitations of tau PET imaging—namely high radiation exposure, high cost, and limited accessibility—in Alzheimer’s disease (AD) diagnosis, this study proposes a novel method for synthesizing high-fidelity tau PET images from T1-weighted MRI. We introduce an innovative VQGAN architecture integrating multi-scale convolution, ResNet-style residual blocks, and the CBAM attention mechanism, enabling end-to-end pathology-aware synthesis. Evaluated on the ADNI dataset (222 paired T1-MRI/tau-PET samples), our method achieves MSE = 0.0056 ± 0.0061, PSNR = 24.39 ± 4.49 dB, and SSIM = 0.9000 ± 0.0453—significantly outperforming cGAN, WGAN-GP, and CycleGAN. Crucially, a downstream AD classifier attains 65.91% accuracy using synthesized tau PET images, closely matching performance on real tau PET (63.64%), thereby validating faithful preservation of pathological features and demonstrating strong potential for clinical substitution.

Technology Category

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

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Tau positron emission tomography (PET) is a critical diagnostic modality for Alzheimer's disease (AD) because it visualizes and quantifies neurofibrillary tangles, a hallmark of AD pathology. However, its widespread clinical adoption is hindered by significant challenges, such as radiation exposure, limited availability, high clinical workload, and substantial financial costs. To overcome these limitations, we propose Multi-scale CBAM Residual Vector Quantized Generative Adversarial Network (MCR-VQGAN) to synthesize high-fidelity tau PET images from structural T1-weighted MRI scans. MCR-VQGAN improves standard VQGAN by integrating three key architectural enhancements: multi-scale convolutions, ResNet blocks, and Convolutional Block Attention Modules (CBAM). Using 222 paired structural T1-weighted MRI and tau PET scans from Alzheimer's Disease Neuroimaging Initiative (ADNI), we trained and compared MCR-VQGAN with cGAN, WGAN-GP, CycleGAN, and VQGAN. Our proposed model achieved superior image synthesis performance across all metrics: MSE of 0.0056 +/- 0.0061, PSNR of 24.39 +/- 4.49 dB, and SSIM of 0.9000 +/- 0.0453. To assess the clinical utility of the synthetic images, we trained and evaluated a CNN-based AD classifier. The classifier achieved comparable accuracy when tested on real (63.64%) and synthetic (65.91%) images. This result indicates that our synthesis process successfully preserves diagnostically relevant features without significant information loss. Our results demonstrate that MCR-VQGAN can offer a reliable and scalable surrogate for conventional tau PET imaging, potentially improving the accessibility and scalability of tau imaging biomarkers for AD research and clinical workflows.

Problem

• Synthesizes tau PET images from MRI to reduce radiation exposure
• Addresses high costs and limited availability of tau PET scans
• Enables scalable Alzheimer's disease imaging with preserved diagnostic features

Innovation

• Multi-scale CBAM Residual VQGAN synthesizes tau PET from MRI
• Integrates multi-scale convolutions, ResNet blocks, and CBAM modules
• Achieves high-fidelity synthesis with superior MSE, PSNR, SSIM metrics