Harmonization of breast MRI data

乳腺 MRI 数据的协调

• 批准号: 10703350
• 负责人: Maciej A. Mazurowski
• 金额: $ 52.4万
• 依托单位: DUKE UNIVERSITY
• 依托单位国家: 美国
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-09-15 至 2026-06-30
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10703350
• 关键词: Abdomen; Address; Adoption; Algorithms; Appearance; Brain; Breast; Breast Magnetic Resonance Imaging; Cancer Detection; Clinical; Collaborations; Computational algorithm; Data; Data Set; Diagnosis; Disease; Evaluation; Genomics; Head; Hip region structure; Image; Institution; Journals; Machine Learning; Magnetic Resonance Imaging; Malignant Neoplasms; Mammary Neoplasms; Manufacturer; Methodology; Methods; Modeling; Modernization; Modification; Noise; Organ; Paper; Patient-Focused Outcomes; Patients; Publications; Reproducibility; Research; Research Personnel; Scanning; Scientist; Source; Structure; Texture; Thinking; Time; Tissues; Training; Translations; Validation; Work; breast imaging; cancer risk; clinical application; clinical implementation; clinical practice; contrast enhanced; convolutional neural network; deep learning; deep learning algorithm; expectation; experience; generative adversarial network; genomic signature; high risk; imaging modality; improved; innovation; linear transformation; novel; novel strategies; patient prognosis; practical application; predictive modeling; radiologist; radiomics; reconstruction; tumor

ABSTRACT Different magnetic resonance imaging (MRI) scanners and different acquisition parameters can produce very different images for the same patients. This is a significant issue when attempting to use MRIs in a quantitative manner. Multiple studies have shown promise of quantitative analysis of breast MRIs to diagnose breast tumors, predict patient outcomes, assess cancer risk, and even identify genomic signatures of cancers. However, the issue of inhomogeneity of images hampers the progress of the research and clinical implementation of these findings. In many cases one cannot utilize images from different sources to answer a research question. Furthermore, predictive models developed at one institution may not generalize to other institutions. While this is a well-recognized problem, there is currently no solution to it in breast MRI. Some valid efforts have been undertaken in order to address this issue for other organs, predominantly brain. However, the problem has not been solved for those organs neither and limited validation of the existing methods in practical contexts hampers the implementation. Breast is a non-rigid organ with highly variable composition making the harmonization of breast MRIs particularly challenging and making almost all prior harmonization methods developed for brain not applicable. Given the urgent need for harmonization in quantitative research, we propose three harmonization methods that allow for transforming an image acquired using one scanner setup to assume appearance of another scanner setup. We introduce important technical innovations to utilize cutting-edge convolutional neural networks for this task. Additionally, we propose a new approach to the question that has not yet attracted significant systematic consideration: what makes a harmonization algorithm successful or useful? We do not evaluate pixel-to-pixel match between the harmonized image and a reference image which is the typical approach. This approach is impractical in breast imaging since it requires ideally paired images, it does not deal well with expected image noise, and it does not inform about specific limitations of the evaluated harmonization method. We propose an evaluation framework that assesses harmonization algorithms in terms of different practical applications including radiomic analysis and deep learning. The study will be conducted in collaboration between a machine learning scientists (Duke and Yale), a breast MRI physicist (Cornell), a radiologist whose research focuses on MRI (Duke), and a biostatistician (Duke). The proposed harmonization and evaluation methods do not require fully paired data and do not make assumptions about tissue composition. Therefore, they will be applicable across other organs once implemented with appropriate data for the organ. All harmonization and evaluation algorithms along with the data will be made publicly available to spearhead further research on this crucial unsolved research topic.

抽象的 不同的磁共振成像（MRI）扫描仪和不同的采集参数可以产生非常非常 同一患者的不同图像。这是尝试在定量中使用MRI时的重要问题 方式。多项研究表明，对乳腺MRI的定量分析有望诊断乳房 肿瘤，预测患者的结局，评估癌症风险，甚至确定癌症的基因组特征。 但是，图像不均匀性的问题阻碍了研究和临床的进步 这些发现的实施。在许多情况下，人们无法利用来自不同来源的图像回答 研究问题。此外，一个机构开发的预测模型可能不会推广到其他机构 机构。尽管这是一个公认的问题，但目前在乳房MRI中尚无解决方案。一些 为了解决其他器官，主要是大脑，已经采取了有效的努力。 但是，对于那些器官尚未解决问题 实际情况下的方法会阻碍实现。乳房是一个非刚性器官，高度可变 构图使乳房MRI的统一特别具有挑战性，并使几乎所有先前 为不适用的大脑开发的协调方法。考虑到迫切需要协调 定量研究，我们提出了三种允许转换获得的图像的协调方法 使用一个扫描仪设置来假设另一个扫描仪设置的外观。我们介绍重要的技术 利用尖端卷积神经网络来完成此任务的创新。此外，我们提出了一个新的 解决尚未引起大量系统考虑的问题的方法：什么使 协调算法成功还是有用？我们不评估像素到像素的匹配 统一图像和参考图像是典型方法。这种方法在乳房中不切实际 成像由于它需要理想的配对图像，因此它不能很好地处理预期图像噪声，并且不会 告知评估协调方法的具体局限性。我们提出了一个评估框架 通过不同的实际应用评估统一算法，包括放射线分析 和深度学习。该研究将通过机器学习科学家（杜克大学）合作进行。 耶鲁大学MRI物理学家（Cornell），一名放射科医生，其研究专注于MRI（DUKE）和A 生物统计学家（公爵）。提出的统一和评估方法不需要完全配对的数据 并且不要对组织组成做出假设。因此，它们将适用于其他器官 一旦实施了适合器官的数据。所有协调和评估算法以及 这些数据将公开使用，以率先对这个至关重要的研究主题进行进一步的研究。