TRD 2 - Deep Learning

TRD 2 - 深度学习

• 批准号: 10090282
• 负责人: William M. Wells
• 金额: $ 28.59万
• 依托单位: BRIGHAM AND WOMEN'S HOSPITAL
• 依托单位国家: 美国
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-01-06 至 2025-12-31
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10090282
• 关键词: Acoustic Neuroma; Address; Brain; Brain Neoplasms; Brain imaging; Cancerous; Classification; Clinical Data; Collaborations; Collection; Communities; Data; Data Set; Disease; Evaluation; Excision; Gleason Grade for Prostate Cancer; Goals; Histopathology; Hospitals; Image; Imaging problem; Informatics; Infrastructure; Label; Lesion; Lung; Lung CAT Scan; Magnetic Resonance Imaging; Malignant Neoplasms; Malignant neoplasm of brain; Malignant neoplasm of lung; Malignant neoplasm of prostate; Manuals; Maps; Medical Imaging; Metadata; Methods; Modeling; Multi-Institutional Clinical Trial; Operative Surgical Procedures; Performance; Pituitary Gland; Probability; Prostate; Psychological Transfer; Readability; Research; Scanning; Supervision; System; Technology; Therapeutic Studies; Training; Ultrasonography; Work; annotation system; anticancer research; base; clinical application; clinically significant; data acquisition; data harmonization; data modeling; deep learning; deep learning algorithm; design; detection method; heterogenous data; image guided; image guided intervention; image guided therapy; image registration; imaging informatics; learning strategy; malignant breast neoplasm; novel; open source; prospective; prostate biopsy; prostate cancer model; success; supervised learning; three dimensional structure; tool

The goals of the Deep Learning TRD of the Advanced Technologies for the National Center for Image-Guided Therapy (AT-NCIGT) are to investigate revolutionary advances in deep learning (DL) in the context of image-guided therapy (IGT) of brain, prostate, and lung cancer, and to develop tools that can be used by the broader IGT research community. The general theme of our research addresses difficulties associated with creation of training data, which is a significant impediment to the application of DL to medical images. While DL has had many successes in image-based classification or segmentation tasks, the methods used are fully supervised, i.e., very large amounts of accurately annotated training data are needed to achieve best performance. Currently, expert annotation is expensive and laborious in the case of medical images because accurate segmentation of 3D structures requires manual or semi-automatic labeling of thousands of voxels per image. Concurrently, large unlabeled or weakly-labeled data sets are becoming available. For example, the PACS system of a large hospital might contain tens of millions of scans, but determining accurate disease labels is difficult. There are currently two promising DL approaches that can be used to address this problem, weakly-supervised learning (where some labels are absent or otherwise imperfect) and transfer learning (which leverages labeled data sets that are in some ways similar). The current situation is further exacerbated by a lack of machine readable metadata, and of methods and tools to support curation of the imaging (e.g., Magnetic Resonance Imaging (MRI)) and clinical data, alongside annotations and analysis results within a single data model. The latter leads to fragmentation of data, and non-standard and heterogeneous metadata. We address these problems by 1) Developing new information theoretic technology for weakly-supervised deep learning, 2) developing novel training strategies for deep learning for cancer characterization for transperineal in-bore MRI-guided prostate biopsy, and 3) developing an infrastructure for curating imaging data for deep learning. The results of this TRD will be DL algorithms, the resulting models, and tools for annotation and organizing machine-readable metadata that are designed to enable IGT cancer research for the prostate, the lung, and the brain applications.

国家先进技术中心深度学习TRD的目标 图像引导治疗（AT-NCIGT）旨在研究深度学习（DL）的革命性进展， 脑、前列腺和肺癌的图像引导治疗（IGT）的背景，并开发工具， 可以被更广泛的IGT研究社区使用。我们研究的总主题是 与创建训练数据有关的困难，这是应用 医学图像。虽然DL在基于图像的分类或分割方面取得了许多成功， 任务，所使用的方法是完全监督，即，大量精确注释的训练 需要数据来实现最佳性能。目前，专家注释是昂贵和费力的， 医学图像的情况下，因为3D结构的准确分割需要手动或 半自动标记每个图像的数千个体素。同时，大的未标记或 弱标记的数据集正在变得可用。例如，大型医院的PACS系统 可能包含数千万次扫描，但很难确定准确的疾病标签。有 目前有两种有前途的DL方法可以用来解决这个问题，弱监督 学习（其中一些标签不存在或不完美）和迁移学习（利用 在某些方面相似的标记数据集）。目前的局势因缺乏 机器可读元数据，以及支持成像管理的方法和工具（例如，磁 磁共振成像（MRI））和临床数据，以及注释和分析结果在一个单一的 数据模型后者导致数据碎片化，以及元数据不标准和异构。 我们解决这些问题，1）发展新的信息理论技术， 弱监督深度学习，2）为癌症深度学习开发新的训练策略 经会阴孔内MRI引导前列腺活检的表征，以及3）开发基础设施 for curating策展imaging成像data数据for deep深learning学习.此TRD的结果将是DL算法， 模型，以及用于注释和组织机器可读元数据的工具， IGT癌症研究在前列腺、肺和脑中的应用。