AAD-DCE: An Aggregated Multimodal Attention Mechanism for Early and Late Dynamic Contrast Enhanced Prostate MRI Synthesis

2025-02-04

AI Summary

To address the toxicity risks of gadolinium-based contrast agents in dynamic contrast-enhanced MRI (DCE-MRI) and the neglect of local perfusion modeling in existing synthesis methods, this paper proposes a multimodal-guided generative adversarial network (GAN) that synthesizes early- and late-phase DCE-MRI from T2-weighted, apparent diffusion coefficient (ADC), and pre-contrast T1-weighted images—enabling noninvasive contrast-free acquisition. Our key innovation is a global–local collaborative aggregation attention discriminator, which for the first time embeds attention maps into the spatially driven generator to precisely localize perfusion characteristics within anatomical structures. The architecture supports cross-architectural generalization, underscoring the critical role of attention integration in temporal perfusion modeling. Evaluated on the ProstateX dataset, our method achieves state-of-the-art performance: +0.64 dB PSNR and +0.0518 SSIM for early-phase synthesis; +0.1 dB PSNR and +0.0424 SSIM for late-phase synthesis; and significantly reduced MAE across all metrics.

Technology Category

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

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems

Abstract

Dynamic Contrast-Enhanced Magnetic Resonance Imaging (DCE-MRI) is a medical imaging technique that plays a crucial role in the detailed visualization and identification of tissue perfusion in abnormal lesions and radiological suggestions for biopsy. However, DCE-MRI involves the administration of a Gadolinium based (Gad) contrast agent, which is associated with a risk of toxicity in the body. Previous deep learning approaches that synthesize DCE-MR images employ unimodal non-contrast or low-dose contrast MRI images lacking focus on the local perfusion information within the anatomy of interest. We propose AAD-DCE, a generative adversarial network (GAN) with an aggregated attention discriminator module consisting of global and local discriminators. The discriminators provide a spatial embedded attention map to drive the generator to synthesize early and late response DCE-MRI images. Our method employs multimodal inputs - T2 weighted (T2W), Apparent Diffusion Coefficient (ADC), and T1 pre-contrast for image synthesis. Extensive comparative and ablation studies on the ProstateX dataset show that our model (i) is agnostic to various generator benchmarks and (ii) outperforms other DCE-MRI synthesis approaches with improvement margins of +0.64 dB PSNR, +0.0518 SSIM, -0.015 MAE for early response and +0.1 dB PSNR, +0.0424 SSIM, -0.021 MAE for late response, and (ii) emphasize the importance of attention ensembling. Our code is available at https://github.com/bhartidivya/AAD-DCE.

Problem

• Synthesizes early and late DCE-MRI images
• Reduces Gadolinium contrast agent toxicity risk
• Improves image quality with multimodal inputs

Innovation

• GAN with aggregated attention discriminator
• Multimodal MRI inputs for synthesis
• Spatial embedded attention map technique