Pooled analysis of multiple sclerosis findings on multi-site 7 Tesla MRI

多部位 7 特斯拉 MRI 多发性硬化症结果的汇总分析

• 批准号: 10430261
• 负责人: Daniel M Harrison
• 金额: $ 48.4万
• 依托单位: UNIVERSITY OF MARYLAND BALTIMORE
• 依托单位国家: 美国
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-07-01 至 2026-06-30
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10430261
• 关键词: Algorithmic Analysis; American; Autopsy; Baltimore; Biological Markers; Caring; Chronic; Clinical; Clinical Data; Clinical Trials; Conflict (Psychology); Consumption; Data; Data Pooling; Demyelinations; Development; Devices; Diagnosis; Early Diagnosis; Evaluation; Functional disorder; Future; Histopathology; Hospitals; Image; Image Analysis; Imaging Techniques; Imaging technology; Impaired cognition; Individual; Inflammation; Inflammatory; Institutes; Institution; Intervention; Lead; Lesion; Link; MRI Scans; Machine Learning; Magnetic Resonance Imaging; Manuals; Maps; Maryland; Masks; Measurement; Measures; Meningeal; Meninges; Methodology; Methods; Monitor; Multiple Sclerosis; Multiple Sclerosis Lesions; National Institute of Neurological Disorders and Stroke; Nature; Neurologic; Outcome; Pathology; Patient Care; Pennsylvania; Persons; Pharmacologic Substance; Phase; Predisposition; Prognosis; Protocols documentation; Research; Research Design; Research Personnel; Research Project Grants; Sample Size; Scanning; Site; Standardization; Structure; Thalamic structure; Time; Training; Universities; Validation; Visit; Visualization; Woman; accurate diagnosis; clinical care; clinically relevant; cognitive disability; comparative; deep learning algorithm; diagnostic tool; disability; effective therapy; follow-up; gray matter; high throughput analysis; imaging study; in vivo; insight; interest; large datasets; machine learning algorithm; magnetic field; multiple sclerosis patient; physically handicapped; prognostic tool; prognostication; screening; tool; treatment effect; white matter

Project Summary/Abstract Background: Although current MRI performs well as a tool to measure white matter (WM) inflammation in MS, its ability to quantify gray matter (GM) pathology, meningeal inflammation, and chronic inflammatory changes are limited. For this reason, researchers have looked to 7-tesla (7T) MRI for more robust in vivo measures of pathology. Although the increased sensitivity of 7T MRI for these aspects of MS have stimulated much interest, the impact of previous studies have been limited due to small sample sizes, varying methodologies, cumbersome analysis methods, and conflicting results – all of which have limited the generalizability of findings and more widespread dissemination of the use of this tool. Objective: In this study, we aim to advance the use of 7T MRI as a tool for diagnosis, prognosis, and treatment effect monitoring by performing multi-site, large sample size evaluation of imaging correlates of cortical and deep GM lesions, chronic active WM lesions (WM lesions with paramagnetic rims on MRI), and meningeal inflammation (leptomeningeal inflammation (LME) on MRI) and validation of their clinical relevance. We also will look develop automated tools for identification and quantification of these findings on 7T MRI. Study Design: Imaging and clinical data will be derived from previous and ongoing studies of the use of 7T MRI in MS performed at multiple collaborative sites under the umbrella of the North American Imaging in MS (NAIMS) Collaborative, including the University of Maryland, Baltimore, Harvard-Brigham and Women’s Hospital, the National Institutes of Neurological Disorders and Stroke, the University of Pennsylvania, and the Montreal Neurological Institute at McGill University. This includes data on up to 814 individual persons with MS, 1232 individual study visits, and a follow up period of up to 5 years. GM lesions, WM lesions with paramagnetic rims, and LME will be identified on 7T, with critical evaluation of methods and their relationship to clinical data. Further, the manual analysis results will be used as training sets for deep learning algorithms for identification of these abnormalities on future 7T MRI scans. Impact: This study would provide the data on measurement variability and generalizable clinical importance of the imaging findings being investigated, along with providing automated tools for future, high-throughput studies. In advancing these precise measures of aspects of MS pathology only poorly measured on lower field MRI, this study will lead to more accurate MS diagnoses, improvements in the ability to monitor changes in MS pathology over time, and perhaps lead to the screening of new pharmaceutical interventions to target GM lesions, WM lesions with paramagnetic rims, and LME.

项目总结/摘要 背景：尽管目前MRI作为测量MS中白色物质（WM）炎症的工具表现良好， 其量化灰质（GM）病理、脑膜炎症和慢性炎症变化的能力 是有限的。出于这个原因，研究人员已经将目光投向了7特斯拉（7 T）MRI，以获得更强大的体内测量。 病理尽管7 T MRI对MS这些方面的敏感性增加， 兴趣，以前的研究的影响是有限的，由于样本量小，不同的方法， 繁琐的分析方法和相互矛盾的结果-所有这些都限制了调查结果的普遍性 并更广泛地传播这一工具的使用。 目的：在这项研究中，我们的目的是提高7 T MRI作为诊断，预后， 通过多部位、大样本量评价影像学相关性进行治疗效果监测， 皮质和深部GM病变，慢性活动性WM病变（MRI上有顺磁性边缘的WM病变），以及 脑膜炎症（MRI上的软脑膜炎症（LME））并验证其临床相关性。 我们还将开发自动化工具，用于识别和量化7 T MRI上的这些发现。 研究设计：影像学和临床数据将来源于既往和正在进行的使用7 T的研究 在MS北美成像的保护下，在多个合作中心进行MS MRI （NAIMS）合作，包括马里兰州，巴尔的摩，哈佛大学布里格姆和妇女的大学 医院，国家神经疾病和中风研究所，宾夕法尼亚大学， 麦吉尔大学蒙特利尔神经学研究所。这包括多达814名个人的数据， MS，1232次个体研究访视，随访期长达5年。GM病变，WM病变伴 顺磁轮辋，和LME将确定在7 T上，与关键的评价方法和它们的关系 临床数据。进一步，人工分析结果将被用作深度学习算法的训练集 用于在未来的7 T MRI扫描中识别这些异常。 影响：本研究将提供关于测量变异性和可推广的临床重要性的数据， 正在研究的成像结果，沿着为未来的高通量 问题研究在推进MS病理学方面的这些精确测量时，仅在低场测量不佳 MRI，这项研究将导致更准确的MS诊断，提高监测MS变化的能力， 随着时间的推移，病理学，也许会导致筛选新的药物干预措施，以靶向转基因 病变、具有顺磁性边缘的WM病变和LME。