Brain phantom generation by generative adversarial net (GAN) for AI-based emission tomography

通过生成对抗网络 (GAN) 生成脑模型，用于基于人工智能的发射断层扫描

• 批准号: 10466967
• 负责人: Wenyi Shao
• 金额: $ 8.19万
• 依托单位: JOHNS HOPKINS UNIVERSITY
• 依托单位国家: 美国
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-08-10 至 2023-05-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/10466967
• 关键词: Abdomen; Address; Affect; Artificial Intelligence; Biological Markers; Brain; Cancer Biology; Cardiac; Cerebrum; Classification; Collimator; Computer Analysis; Computer Vision Systems; Corpus striatum structure; Data; Development; Diagnosis; Discipline of Nuclear Medicine; Disease; Educational Status; Effectiveness; Future; Generations; Goals; Human; Image; Imaging Techniques; Intervention; Investigation; Label; Learning; Limb structure; Machine Learning; Magnetic Resonance Imaging; Maps; Medical; Medical Imaging; Medical Research; Methods; Monitor; Morphologic artifacts; Nerve Degeneration; Noise; Organ; Output; Parkinson Disease; Patients; Performance; Perfusion; Persons; Physics; Population; Positron-Emission Tomography; Procedures; Publishing; Research; Resolution; Scheme; Speed; System; Time; Training; Training Support; Transplantation; United States; Vascular Diseases; Water; Width; Work; attenuation; base; bone; cancer imaging; data access; design; digital; digital imaging; dopaminergic neuron; generative adversarial network; heart imaging; image guided; image reconstruction; improved; interest; nervous system disorder; neural network; neurotransmission; nigrostriatal degeneration; putamen; radiotracer; reconstruction; reduce symptoms; simulation; single photon emission computed tomography; tomography; tool; uptake; virtual

Project Summary Positron emission tomography (PET) and single photon emission computed tomography (SPECT) are useful functional medical imaging techniques that can be performed to evaluate brain functions such as regional cerebral perfusion and neurotransmission. The spatial resolution of reconstruction for PET is usually 3-6 mm, and for SPECT is only 1-2 cm. Motivated by the latest advances in artificial intelligence (AI)/machine learning (ML) and its successful application to MRI and CT, it is highly desirable to develop an ML-based system for PET/SPECT cerebral image reconstruction (one of our specific interests is Parkinson disease) to achieve higher resolution and lower noise than using conventional approaches. However, to develop such a learning system, ground-truth data (accurate images, used as the labels) that guide the training are unavailable from the the real world. Published ML systems for PET imaging have used reconstructed images from conventional methods as the label to guide the training. As a result, the goal was only targeted to improve reconstruction speed, rather than improving the image quality. Since the quality of reconstructed image by ML system cannot exceed the guiding images, the performance of ML system cannot surpass conventional methods. Therefore, in this project, we propose a two-year project that will use conditional generative adversarial networks (GAN) to generate digital 2-D human brain phantoms, which will be highly similar to real human brains. The generated phantoms will serve as the (precise) ground-truth data to develop ML-based PET/SPECT reconstruction systems (our future research). The generated phantoms will contain an activity image and an attenuation map. Hence, results from this work can be used for simulating brain PET or SPECT examinations for various neurological disorders, and neural network can be trained with known ground truth. In addition, designing ML systems often relies on large amounts of data, but it is not easy to access data from a large number of patients in the US for specific medical research (mature ML systems developed for computer vision and image classification often involve images on the million level for training). Existing ML systems developed for MRI, CT, and PET imaging often merely uses a few tens of patient data for training and even less data to validate. Therefore, those systems are high-likely overfitted to the data used in training. With the generation system proposed from this project, we can produce a large phantom population to avoid the overfitting problem when design the AI image-reconstruction system. Once the GAN system is successfully developed, it can be easily transplanted to phantom generation for the AI-based CT and AI-based MRI. The method is also potentially extendable to generate phantom populations of torso, abdomen, and extremities for simulating cardiac imaging, tumor imaging, etc.

项目摘要 正电子发射断层扫描（PET）和单光子发射计算机断层扫描（SPECT）是有用的 功能性医学成像技术，可以用来评估大脑的功能，如局部 脑灌注和神经传递。PET重建的空间分辨率通常为3-6 mm， 而SPECT仅为1-2 cm。受人工智能（AI）/机器学习最新进展的启发 (ML)及其在MRI和CT中的成功应用，非常需要开发一种基于ML的系统， PET/SPECT脑图像重建（我们的具体兴趣之一是帕金森病），以实现更高的 分辨率和更低的噪声比使用传统的方法。然而，为了开发这样的学习系统， 从真实的中无法获得指导训练的地面实况数据（准确的图像，用作标签） 世界用于PET成像的公开的ML系统已经使用来自常规方法的重建图像， 标签来指导培训。因此，目标只是针对提高重建速度，而不是 而不是提高图像质量。由于ML系统重建图像的质量不能超过 引导图像，ML系统的性能不能超过传统的方法。因此，在本项目中， 我们提出了一个为期两年的项目，将使用条件生成对抗网络（GAN）， 生成与真实的人脑高度相似的数字化二维人脑模型。的 生成的体模将作为（精确的）地面实况数据来开发基于ML的PET/SPECT 重建系统（我们未来的研究）。生成的幻影将包含活动图像和 衰减图因此，本工作的结果可用于模拟脑部PET或SPECT检查 对于各种神经系统疾病，神经网络可以用已知的基础事实进行训练。此外，本发明还提供了一种方法， 设计ML系统通常依赖于大量数据，但从大量数据中访问数据并不容易。 在美国的患者进行特定的医学研究（为计算机视觉和图像开发的成熟ML系统） 分类通常涉及用于训练的百万级别的图像）。为MRI、CT、 PET成像通常仅使用几十个患者数据进行训练，甚至更少的数据进行验证。 因此，这些系统很可能过度拟合训练中使用的数据。随着发电系统 从这个项目提出，我们可以产生一个大的幻影人口，以避免过度拟合问题 在设计人工智能图像重建系统时。一旦GAN系统开发成功， 易于移植到基于AI的CT和基于AI的MRI的体模生成中。该方法还可能 可扩展以生成躯干、腹部和四肢的体模群体，用于模拟心脏成像， 肿瘤成像等。

项目成果

Dielectric Breast Phantoms by Generative Adversarial Network.

• DOI: 10.1109/tap.2021.3121149
• 发表时间: 2022-08
• 期刊: IEEE TRANSACTIONS ON ANTENNAS AND PROPAGATION
• 影响因子: 5.7
• 作者: Shao, Wenyi;Zhou, Beibei
• 通讯作者: Zhou, Beibei

Generation of Digital Brain Phantom for Machine Learning Application of Dopamine Transporter Radionuclide Imaging.

• DOI: 10.3390/diagnostics12081945
• 发表时间: 2022-08-12
• 期刊: DIAGNOSTICS
• 影响因子: 3.6
• 作者: Shao, Wenyi;Leung, Kevin H.;Xu, Jingyan;Coughlin, Jennifer M.;Pomper, Martin G.;Du, Yong
• 通讯作者: Du, Yong

Near-Field Microwave Scattering Formulation by A Deep Learning Method.

• DOI: 10.1109/tmtt.2022.3184331
• 发表时间: 2022-11
• 期刊: IEEE TRANSACTIONS ON MICROWAVE THEORY AND TECHNIQUES
• 影响因子: 4.3
• 作者: Shao, Wenyi;Zhou, Beibei
• 通讯作者: Zhou, Beibei