Robust and Efficient Learning of High-Resolution Brain MRI Reconstruction from Small Referenceless Data

从小型无参考数据中稳健而高效地学习高分辨率脑 MRI 重建

• 批准号: 10584324
• 负责人: Mehmet Akcakaya
• 金额: $ 53.06万
• 依托单位: UNIVERSITY OF MINNESOTA
• 依托单位国家: 美国
• 财政年份: 2023
• 资助国家: 美国
• 起止时间: 2023-03-15 至 2027-02-28
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10584324
• 关键词: 3-Dimensional; Acceleration; Affect; Algorithms; Anatomy; Area; Artificial Intelligence; Attention; BRAIN initiative; Behavior Disorders; Behavioral; Brain; Brain Mapping; Brain imaging; Brain scan; Clinical Protocols; Data; Data Set; Databases; Development; Diagnosis; Diffusion; Diffusion Magnetic Resonance Imaging; Dimensions; Disease; Evaluation; Face; Four-dimensional; Functional Magnetic Resonance Imaging; Future; Health Care Costs; Healthcare; Healthcare Systems; Human; Image; Individual; Learning; MRI Scans; Magnetic Resonance Imaging; Maps; Measures; Memory; Mental disorders; Methodology; Methods; Minnesota; Modeling; Morphologic artifacts; Neurologic; Neurons; Pathology; Patients; Performance; Physics; Play; Process; Protocols documentation; Psyche structure; Recovery; Research; Resolution; Role; Sampling; Scanning; Series; Speed; Structure; Supervision; Symptoms; Techniques; Technology; Time; Training; Translations; Uncertainty; United States; Universities; Validation; Work; brain magnetic resonance imaging; connectome; data space; deep learning; denoising; diagnostic value; disability; disability-adjusted life years; image processing; image reconstruction; imaging modality; improved; learning strategy; millimeter; nervous system disorder; neural network; neuropsychiatry; novel; palliative; rapid technique; reconstruction; spatiotemporal; supervised learning; technology development; theories; tool; years of life lost

PROJECT SUMMARY/ABSTRACT Neuropsychiatric (mental, behavioral and neurological) disorders are increasingly dominating the burden on US healthcare. Yet, our understanding of such disorders is largely restricted to a description of symptoms, and the treatments remain palliative. Several large-scale efforts, including the Human Connectome Project (HCP) and the BRAIN Initiative call for the development of technologies to map brain circuits to improve our understanding of brain function. Magnetic resonance imaging (MRI) plays a central role in these initiatives as a powerful non-invasive methodology to study the human brain, including anatomical, functional and diffusion imaging. Yet, MRI methods have major limitations on achievable resolutions and acquisition speed. These affect both high resolution whole brain acquisitions that aim to image voxel volumes that contain only a few thousand neurons for improved understanding of the brain, and also the more commonly utilized research and clinical protocols. This, in turn, necessitates improved reconstruction methods to facilitate faster acquisitions. Several strategies have been proposed for improved reconstruction of MRI data. Recently, deep learning (DL) has emerged as an alternative for accelerated MRI showing improved quality over conventional approaches. However, it also faces challenges that hinder its utility, especially in high-resolution brain MRI, including need for large databases of reference data for training, concerns about generalization to unseen pathologies not well-represented in training datasets, robustness issues related to recovery of fine structures, and difficulties in training networks for processing multi-dimensional image series. In this proposal, we will develop and validate robust and efficient learning strategies for high-resolution brain DL MRI reconstruction without large databases of reference data. We will develop self-supervised learning methods for training with small referenceless databases or in a scan-specific manner. We will augment these with uncertainty-guided training strategies for improved recovery of areas with high uncertainty, methods for synergistically combining random matrix theory based denoising with DL reconstruction, and memory-efficient distributed learning techniques to process large image series. Our developments will enable at least a two-fold improvement in acceleration rates over existing protocols, and at higher resolutions. They will be validated on HCP-style acquisitions with extensive anatomical, functional and microstructural evaluation at multiple resolutions. Finally, we will curate a whole brain sub-millimeter HCP-style database for studying functional and structural connectivity at the level cortical layers and columns, while also facilitating technical developments for new modeling, image processing and reconstruction algorithms. Successful completion of this project has the potential to transform the scales that can be imaged with MRI, improve the quality of existing protocols and/or significantly reduce scan times, leading to reductions in healthcare costs, improved diagnosis and/or increased patient throughput.

项目概要/摘要 神经精神（精神、行为和神经）疾病日益成为人们的负担 美国医疗保健。然而，我们对此类疾病的理解很大程度上仅限于症状的描述，并且 治疗仍然是姑息治疗。多项大规模工作，包括人类连接组计划 (HCP) BRAIN Initiative 呼吁开发绘制大脑回路的技术，以改善我们的大脑 了解大脑功能。磁共振成像 (MRI) 在这些举措中发挥着核心作用 强大的非侵入性方法来研究人脑，包括解剖学、功能和扩散 成像。然而，MRI 方法在可实现的分辨率和采集速度方面存在重大限制。这些 影响高分辨率全脑采集，其目的是对仅包含少数的体素体积进行成像 数千个神经元，以提高对大脑的理解，以及更常用的研究和 临床方案。反过来，这需要改进重建方法以促进更快的采集。 已经提出了几种改进 MRI 数据重建的策略。最近，深度学习（DL） 已成为加速 MRI 的替代方案，显示出比传统方法更高的质量。 然而，它也面临着阻碍其实用性的挑战，特别是在高分辨率脑部 MRI 方面，包括需要 对于用于训练的大型参考数据数据库，不关心对看不见的病理的泛化 在训练数据集中得到很好的体现，与精细结构恢复相关的鲁棒性问题以及 用于处理多维图像系列的训练网络。在此提案中，我们将开发并验证 无需大型数据库即可实现高分辨率大脑 DL MRI 重建的稳健且高效的学习策略 的参考数据。我们将开发自我监督学习方法，用于小无参考的训练 数据库或以扫描特定的方式。我们将通过不确定性引导的培训策略来增强这些 改进高不确定性区域的恢复，协同结合随机矩阵理论的方法 基于深度学习重建的去噪，以及内存高效的分布式学习技术来处理大数据 图像系列。我们的开发将使加速度比现有技术至少提高两倍 协议，并且分辨率更高。它们将在 HCP 式的收购中得到广泛的验证 多种分辨率下的解剖、功能和微观结构评估。最后我们会整理出一个完整的 大脑亚毫米 HCP 型数据库，用于研究皮质水平的功能和结构连接 层和列，同时还促进新建模、图像处理和 重建算法。该项目的成功完成有可能改变规模 可以通过 MRI 成像，提高现有协议的质量和/或显着减少扫描时间， 从而降低医疗成本、改善诊断和/或增加患者吞吐量。