DL-based CT image formation with characterization and control of resolution and noise

基于深度学习的 CT 图像形成，具有分辨率和噪声的表征和控制

基本信息

批准号：
10666105
负责人：
JINGYAN XU
金额：
$ 25.66万
依托单位：
JOHNS HOPKINS UNIVERSITY
依托单位国家：
美国
项目类别：
财政年份：
2023
资助国家：
美国
起止时间：
2023-09-01 至 2026-07-31
项目状态：
未结题

来源：
https://reporter.nih.gov/project-details/10666105
关键词：
Affect Calcium Cardiac Cardiac Catheterization Procedures Clinic Clinical Computer Vision Systems Data Data Collection Daughter Detection Development Diagnosis Ensure Evaluation Excision Future Image Image Analysis Joints Kidney Label Learning Lesion Libraries Liver Methods Modeling Modernization Mothers Names Noise Output Patients Performance Photons Physics Process Proliferating Property Research Research Personnel Resolution Running Sampling Signal Transduction Site Specific qualifier value System Systems Analysis Task Performances Technology Testing Time Training Validation Work X-Ray Computed Tomography X-Ray Medical Imaging abdominal CT accurate diagnosis calcification clinical application clinical efficacy clinical translation coronary computed tomography angiography deep learning deep neural network design digital experience flexibility follow-up image guided image reconstruction imaging modality imaging properties improved innovation interest learning network loss of function neural network architecture prospective radiologist response statistics

项目摘要

Deep learning (DL) based CT image formation methods have proliferated over the past few years. The existing approaches mostly follow the paradigm established in computer vision, and build a deep neural network (DNN) with standard modules that capture salient image features useful for computer vision tasks. These standard modules also work very well for CT images, placing DL-based CT image formation methods at the forefront of research and innovation. However, current DNNs are oblivious to the fact that CT images, unlike natural images, must be interpretable by a radiologist to make a diagnosis. CT image interpretation is affected by image features such as image resolution and noise variance-covariance, which are under exploited by the standard modules from computer vision. Consequently, current DL-based CT image formation has no direct characterization, let alone prospective control, of image resolution and noise variance-covariance. These properties can only be assessed after an image is generated, but resolution/noise has no direct influence during the image formation process. In this proposal, we challenge this established paradigm and propose an innovative DL framework named GradDNN to (1) characterize the resolution and noise properties of a DNN’s output during network training or parameter fine-tuning, and to (2) guide the image formation process so that the output has the desired resolution/noise properties. GradDNN (which stands for gradient + DNN) applies network linearization, i.e., gradient computation, to a mother DNN to extract local resolution and noise properties of the images generated by the mother, and make these properties available during network training and parameter fine tuning. The linearization method for analyzing nonlinear systems such as a DNN has never been attempted before. Conceptually, GradDNN associates a daughter module to any mother DNN for noise/resolution characterization, thereby making the resulting network CT-specific. We will develop GradDNN and demonstrate its capability in the context of two important clinical tasks: (1) mitigation of calcium blooming in coronary CT angiography, and (2) low contrast lesion detection in abdominal CT, both of which have high requirements on resolution and noise. Data for DL network training will be generated using digitally augmented patient data prospectively collected at two sites. Image quality comparison between the mother DNN alone and the mother+daughter duo, using a number of effective mother DNN architectures, will be carried out to demonstrate the additional gain of DL with joint resolution/noise learning. The comparison will use both digitally augmented patient data and real patient data to further establish robustness and generalizability. In this exploratory proposal, we focus on DL networks that perform image-to-image transformation. However, the learning framework using GradDNN is general and can be applied to DL networks that perform direct projection-to-image transformation. Successful completion of this proposal will generate strong preliminary data for a follow-up R01 that extends our results to such DL networks.

基于深度学习 (DL) 的 CT 图像形成方法在过去几年中激增。现有的方法大多遵循计算机视觉中建立的范式，并构建深度神经网络网络（DNN），具有标准模块，可捕获对计算机视觉任务有用的显着图像特征。这些标准模块也非常适合 CT 图像，将基于 DL 的 CT 图像形成方法置于然而，当前的 DNN 忽视了 CT 图像、 CT 图像解读与自然图像不同，必须由放射科医生解读才能做出诊断。受图像分辨率和噪声方差-协方差等图像特征的影响，这些特征尚未得到充分利用通过计算机视觉的标准模块检查，当前基于深度学习的 CT 图像形成没有。图像分辨率和噪声方差-协方差的直接表征，更不用说前瞻性控制。属性只能在生成图像后评估，但分辨率/噪声没有直接影响在这个提案中，我们挑战了这个既定的范式并提出了一个。名为 GradDNN 的创新深度学习框架 (1) 表征 DNN 的分辨率和噪声属性网络训练或参数微调期间的输出，并（2）指导图像形成过程，以便输出具有所需的分辨率/噪声属性。GradDNN（代表梯度 + DNN）适用。对母 DNN 进行网络线性化（即梯度计算）以提取局部分辨率和噪声母亲生成的图像的属性，并在网络训练期间使这些属性可用用于分析 DNN 等非线性系统的线性化方法从未出现过。从概念上讲，GradDNN 将子模块与任何母 DNN 相关联。噪声/分辨率表征，使生成的网络具有 CT 特定性，我们将由此开发 GradDNN。并在两个重要的临床任务中展示其能力：（1）减轻钙华冠状动脉 CT 血管造影中的低对比度病变检测，以及（2）腹部 CT 中的低对比度病变检测，两者都具有高深度学习网络训练的数据将使用数字增强技术生成。在两个地点前瞻性收集的患者数据与单独的母亲 DNN 之间的图像质量比较。母女二人组将使用许多有效的母 DNN 架构展示 DL 与联合分辨率/噪声学习的额外增益该比较将使用两者的数字方式。增强了患者数据和真实患者数据，以进一步建立稳健性和普遍性。在探索性提案中，我们专注于执行图像到图像转换的深度学习网络。使用 GradDNN 的学习框架是通用的，可以应用于执行直接操作的 DL 网络投影到图像的转换的成功完成将产生强有力的初步结果。后续 R01 的数据将我们的结果扩展到此类 DL 网络。