TR&D2: Advanced Statistical Image Reconstruction & Physics Informed Artificial Intelligence for Quantitative PET/MR

TR

• 批准号: 10651773
• 负责人: Quanzheng Li
• 金额: $ 28.08万
• 依托单位: MASSACHUSETTS GENERAL HOSPITAL
• 依托单位国家: 美国
• 财政年份: 2017
• 资助国家: 美国
• 起止时间: 2017-09-30 至 2027-04-30
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10651773
• 关键词: 18F-fluorothymidine; 3-Dimensional; Acceleration; Address; Algorithms; Anatomy; Area; Artificial Intelligence; Cephalic; Clinical; Data; Data Set; Development; Diagnostic; Disease; Disease Progression; Dose; Early treatment; Environment; Evaluation; FOLH1 gene; Funding; Goals; Human; Image; Imaging technology; Intervention; Knowledge; Lead; Learning; Longitudinal Studies; Magnetic Resonance Imaging; Markov Chains; Medical; Medical Imaging; Medicine; Methodology; Methods; Modeling; Motion; Network-based; Noise; Organ; Physics; Positron-Emission Tomography; Protocols documentation; Research; Scanning; Signal Transduction; Solid; Source; Standardization; Technology; Time; Tracer; Training; Uncertainty; accurate diagnostics; anatomic imaging; autoencoder; data analysis pipeline; deep learning; deep neural network; denoising; experience; fluorodeoxyglucose; fluorodeoxyglucose positron emission tomography; image reconstruction; image translation; imaging modality; improved; innovation; learning strategy; molecular imaging; novel; parametric imaging; precision medicine; quantitative imaging; radiotracer; rate of change; reconstruction; response; simulation; success; tau Proteins; technology development; transfer learning; treatment response; ultra high resolution; uptake

The development of artificial intelligence (AI) methodology is of profound importance and is expected to have major societal impact, especially its effect on medicine. In the past funding cycle, we pioneered the application of deep neural networks (DNN) in various image reconstruction tasks and built a solid understanding and extensive experience of its applications in medical imaging. In this new TR&D, we propose use deep learning (DL) to push the application of AI in medical imaging beyond the traditional image reconstruction problem. The study of novel contrast mechanisms (e.g. new MRI sequence and new PET tracer) is a major frontline of PET/MR innovation. To achieve improved image quality, we will incorporate anatomic image and motion correction in a novel DL-based image reconstruction framework. We will also build our AI model based on the accumulated big imaging data at MGH while also providing a methodology to transfer this knowledge to new studies with few existing data. Our proposed domain adaptation and domain adaptation few shot learning technology will largely address some of the biggest challenges of AI in medical imaging, i.e., limited training data, the generalizability problem, thus enabling AI to be practically disseminated and used in clinical environment. Finally, we propose to estimate the posterior distribution of the reconstructed image. The availability of the uncertainty of reconstruction will open a new window for much more elegant and accurate diagnostic protocols and early treatment response evaluation in precision medicine thus leading to a significant number of new applications for PET/MR.

人工智能（AI）方法论的发展具有深远的重要性， 对社会产生重大影响，特别是对医学的影响。在过去的融资周期中， 深度神经网络（DNN）在各种图像重建任务中的应用，并建立了坚实的 了解和丰富的经验，其在医学成像中的应用。在这个新的TR&D中，我们 建议使用深度学习（DL）来推动AI在医学成像中的应用超越传统 图像重建问题。新的对比机制的研究（例如新的MRI序列和 新PET示踪剂）是PET/MR创新的主要前沿。为了提高图像质量，我们 将在基于DL的新型图像重建中结合解剖图像和运动校正 框架.我们还将根据MGH积累的大成像数据构建我们的AI模型， 还提供了一种方法，将这些知识转移到现有数据很少的新研究中。我们 提出的领域自适应和领域自适应少拍学习技术将在很大程度上解决 人工智能在医学成像中的一些最大挑战，即，有限的训练数据， 问题，从而使AI能够在临床环境中实际传播和使用。最后我们 建议估计重建图像的后验分布。的可用性 重建的不确定性将为更优雅和准确的诊断打开新的窗口 方案和早期治疗反应评估，从而导致一个显着的 PET/MR的新应用数量。