MRIQT: Physics-Aware Diffusion Model for Image Quality Transfer in Neonatal Ultra-Low-Field MRI

2025-11-17

AI Summary

To address the low signal-to-noise ratio (SNR) and insufficient diagnostic quality of portable ultra-low-field MRI (uLF-MRI, 0.064 T) for neonates, this paper proposes MRIQT—the first physics-aware 3D diffusion model for uLF-to-high-field (HF) image quality translation (IQT). Methodologically, MRIQT integrates realistic k-space degradation modeling, v-prediction training, classifier-free guidance, and an SNR-weighted 3D perceptual loss to ensure stable, high-fidelity reconstruction while preserving anatomical accuracy. Its volumetric attention U-Net architecture jointly models 3D structural coherence and physical consistency. Evaluated on a clinical neonatal cohort, MRIQT achieves a 15.3% PSNR improvement over state-of-the-art GAN- and CNN-based methods; 85% of its outputs are rated by radiologists as high-quality with clear lesion delineation. This advancement significantly facilitates the clinical deployment of uLF-MRI for bedside neuroimaging assessment in neonates.

Technology Category

• Machine Learning: Quantum Machine Learning
• Computer Vision: Low Level & Physics-based Vision
• Knowledge Representation and Reasoning: Qualitative Reasoning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Portable ultra-low-field MRI (uLF-MRI, 0.064 T) offers accessible neuroimaging for neonatal care but suffers from low signal-to-noise ratio and poor diagnostic quality compared to high-field (HF) MRI. We propose MRIQT, a 3D conditional diffusion framework for image quality transfer (IQT) from uLF to HF MRI. MRIQT combines realistic K-space degradation for physics-consistent uLF simulation, v-prediction with classifier-free guidance for stable image-to-image generation, and an SNR-weighted 3D perceptual loss for anatomical fidelity. The model denoises from a noised uLF input conditioned on the same scan, leveraging volumetric attention-UNet architecture for structure-preserving translation. Trained on a neonatal cohort with diverse pathologies, MRIQT surpasses recent GAN and CNN baselines in PSNR 15.3% with 1.78% over the state of the art, while physicians rated 85% of its outputs as good quality with clear pathology present. MRIQT enables high-fidelity, diffusion-based enhancement of portable ultra-low-field (uLF) MRI for deliable neonatal brain assessment.

Problem

• Enhancing diagnostic quality of neonatal ultra-low-field MRI scans
• Translating low-field MRI to high-field quality using diffusion models
• Improving signal-to-noise ratio while preserving anatomical fidelity

Innovation

• Physics-aware diffusion model for MRI enhancement
• Volumetric attention-UNet for structure preservation
• SNR-weighted perceptual loss for anatomical fidelity