Susceptibility Distortion Correction of Diffusion MRI with a single Phase-Encoding Direction

2025-08-18

AI Summary

Susceptibility-induced geometric and intensity distortions in diffusion MRI (dMRI) pose significant challenges for quantitative analysis, particularly in retrospective single-phase-encoding-direction acquisitions where conventional correction methods—such as FSL’s topup—require complementary blip-up/blip-down data pairs. This work introduces the first end-to-end deep learning framework for susceptibility distortion correction using only a single-polarity dMRI acquisition. Leveraging physics-informed loss functions derived from the dMRI forward model, the method operates without paired ground-truth data or explicit field-map estimation. Evaluated on public benchmarks, it achieves geometric correction accuracy comparable to topup while substantially improving structural fidelity in unidirectional dMRI. By eliminating the need for bipolar acquisitions, this unsupervised, deployable solution overcomes a fundamental limitation of traditional approaches and enables high-quality post-processing of clinical retrospective dMRI data.

Technology Category

• Machine Learning: Multimodal Learning
• Constraint Satisfaction and Optimization: Constraint Learning and Acquisition
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Diffusion MRI (dMRI) is a valuable tool to map brain microstructure and connectivity by analyzing water molecule diffusion in tissue. However, acquiring dMRI data requires to capture multiple 3D brain volumes in a short time, often leading to trade-offs in image quality. One challenging artifact is susceptibility-induced distortion, which introduces significant geometric and intensity deformations. Traditional correction methods, such as topup, rely on having access to blip-up and blip-down image pairs, limiting their applicability to retrospective data acquired with a single phase encoding direction. In this work, we propose a deep learning-based approach to correct susceptibility distortions using only a single acquisition (either blip-up or blip-down), eliminating the need for paired acquisitions. Experimental results show that our method achieves performance comparable to topup, demonstrating its potential as an efficient and practical alternative for susceptibility distortion correction in dMRI.

Problem

• Corrects susceptibility distortions in diffusion MRI
• Uses single phase-encoding direction data only
• Eliminates need for paired blip-up/down acquisitions

Innovation

• Deep learning corrects distortion single direction
• Eliminates need for paired phase encoding
• Performance comparable traditional topup method