From Low Field to High Value: Robust Cortical Mapping from Low-Field MRI

2025-05-18

AI Summary

Existing cortical surface analysis tools exhibit significant performance degradation on low-field MRI (LF-MRI) due to inherently low signal-to-noise ratio and poor spatial resolution. This work introduces the first plug-and-play, zero-shot 3D cortical reconstruction and morphometric framework specifically designed for LF-MRI—requiring no retraining for new datasets and directly accommodating multi-contrast, multi-resolution LF-MRI acquisitions. Our key innovations are: (1) the first use of synthetically generated LF-MRI data to train a 3D U-Net for robust signed distance function (SDF) prediction, enabling geometry-aware surface modeling; and (2) a topology-preserving geometric optimization step that enforces biologically plausible cortical surface topology. Evaluated on 3-mm isotropic T2-weighted LF-MRI scans (acquisition time <4 minutes), the method achieves cortical surface area correlation *r* = 0.96, regional segmentation Dice score = 0.98, and gray matter volume correlation *r* = 0.93. Critically, it generalizes successfully to challenging postmortem LF-MRI data.

Technology Category

• Computer Vision: 3D Computer Vision
• Machine Learning: Large Multimodal Models (LMMs)
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• Economics, Online Markets and Human Computation: Architectures and workflows that use LLMs for crowd work
• Web Mining and Content Analysis: Large pretrained models with web data
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Three-dimensional reconstruction of cortical surfaces from MRI for morphometric analysis is fundamental for understanding brain structure. While high-field MRI (HF-MRI) is standard in research and clinical settings, its limited availability hinders widespread use. Low-field MRI (LF-MRI), particularly portable systems, offers a cost-effective and accessible alternative. However, existing cortical surface analysis tools are optimized for high-resolution HF-MRI and struggle with the lower signal-to-noise ratio and resolution of LF-MRI. In this work, we present a machine learning method for 3D reconstruction and analysis of portable LF-MRI across a range of contrasts and resolutions. Our method works"out of the box"without retraining. It uses a 3D U-Net trained on synthetic LF-MRI to predict signed distance functions of cortical surfaces, followed by geometric processing to ensure topological accuracy. We evaluate our method using paired HF/LF-MRI scans of the same subjects, showing that LF-MRI surface reconstruction accuracy depends on acquisition parameters, including contrast type (T1 vs T2), orientation (axial vs isotropic), and resolution. A 3mm isotropic T2-weighted scan acquired in under 4 minutes, yields strong agreement with HF-derived surfaces: surface area correlates at r=0.96, cortical parcellations reach Dice=0.98, and gray matter volume achieves r=0.93. Cortical thickness remains more challenging with correlations up to r=0.70, reflecting the difficulty of sub-mm precision with 3mm voxels. We further validate our method on challenging postmortem LF-MRI, demonstrating its robustness. Our method represents a step toward enabling cortical surface analysis on portable LF-MRI. Code is available at https://surfer.nmr.mgh.harvard.edu/fswiki/ReconAny

Problem

• Develops 3D cortical surface reconstruction from low-field MRI
• Addresses poor performance of existing tools on low-resolution MRI
• Enables portable MRI analysis without retraining using synthetic data

Innovation

• Machine learning for 3D LF-MRI reconstruction
• Uses synthetic LF-MRI-trained 3D U-Net
• Geometric processing ensures topological accuracy