Next Generation Brain PET Imaging

下一代脑 PET 成像

基本信息

批准号：
10478939
负责人：
Gregory George Zaharchuk
金额：
$ 54.48万
依托单位：
STANFORD UNIVERSITY
依托单位国家：
美国
项目类别：
财政年份：
2021
资助国家：
美国
起止时间：
2021-09-01 至 2026-08-31
项目状态：
未结题

来源：
https://reporter.nih.gov/project-details/10478939
关键词：
Age Air Architecture Artificial Intelligence Brain Brain imaging Cerebrovascular Circulation Characteristics Clinical Clinical Data Clinical Trials Cyclotrons Data Data Set Dementia Diagnosis Diagnostic Diagnostic Imaging Discipline of Nuclear Medicine Disease Disease Progression Dose Epilepsy FDA approved Functional Magnetic Resonance Imaging Future Gender Genetic Genetic Markers Glioblastoma Goals Gold Human Image Link Logistics Magnetic Resonance Imaging Maps Measurement Measures Medical Medical Genetics Medical Imaging Metabolic Metabolism Methods Modality Modeling Patients Performance Phenotype Physicians Population Positron-Emission Tomography Prevalence Protocols documentation Provider Radiation Radiation Dose Unit Radiology Specialty Reader Recording of previous events Recurrence Risk Risk Factors Running Rural Community Site Specific qualifier value Technology Tracer Training Travel Water acute stroke arterial spin labeling base chronic stroke convolutional neural network cost deep learning disease natural history fluorodeoxyglucose fluorodeoxyglucose positron emission tomography genetic information imaging biomarker imaging modality imaging study improved machine learning model metabolic imaging multimodal data multimodality neural network next generation patient population patient safety prospective quantitative imaging radiologist radiotracer sex stroke patient targeted imaging tumor underserved community

项目摘要

Abstract Gold-standard quantitative imaging studies are often difficult to implement, limited by financial and logistical issues, or expose the patient to unnecessary risks. Deep learning has shown great promise in recent years for many medical applications; one use is to synthesize improved images. Such image trans- formation methods offer the potential to improve the quality, value, and accessibility of medical imaging. The goal of this project is to develop deep convolutional neural network approaches to FDG PET imaging, the most commonly performed clinical brain- focused PET study in the USA. Using simultaneous PET/MRI, we will train networks to produce diagnostic PET images from ultra-low dose PET and MR images. We will explore the three reimbursed clinical indications for this imaging modality (tumor recurrence, dementia, and epilepsy) using both quantitative metrics and repeated reader studies to assess equivalence and evaluate possible AI generalization bias related to simultaneity, scanner type, age, gender, and disease prevalence. Next, we will evaluate whether we can move beyond ultra-low dose and remove the radiation dose altogether, synthesizing FDG brain PET images from MR inputs only, relying on the information in multi-modal functional MRI. Finally, we will assess whether we can use deep networks to combine imaging and non- imaging data such as clinical and genetic information to further improve image transformation and predict future images and image-based biomarkers. Significantly reducing or even eliminating the need for radiation to produce brain FDG PET images would be truly transformative while the ability to predict the future will enable personalized radiology and enhance our ability to perform clinical trials.

抽象的黄金标准定量成像研究通常难以实施，受到财务和后勤问题的限制，或使患者面临不必要的风险风险。近年来，深度学习在许多医学领域显示出巨大的前景应用程序；一种用途是合成改进的图像。这样的图像传输形成方法提供了提高质量、价值和医学成像的可及性。该项目的目标是开发深度卷积神经网络 FDG PET 成像方法是临床最常用的脑成像方法美国的重点 PET 研究。使用同步 PET/MRI，我们将训练利用超低剂量 PET 和 MR 生成诊断 PET 图像的网络图像。我们将探讨这三个可报销的临床适应症使用两者的成像方式（肿瘤复发、痴呆和癫痫）定量指标和重复的读者研究来评估等效性和评估与同时性、扫描仪类型相关的可能的人工智能泛化偏差，年龄、性别和疾病患病率。接下来，我们将评估是否可以超越超低剂量并消除综合辐射剂量，从 MR 合成 FDG 脑部 PET 图像仅输入，依赖于多模态功能 MRI 中的信息。最后，我们将评估是否可以使用深度网络将成像和非临床和遗传信息等影像数据进一步改善图像转换和预测未来图像和基于图像的生物标记。显着减少甚至消除产生辐射的需要大脑 FDG PET 图像将真正具有变革性，同时能够预测未来将实现个性化放射学并增强我们的能力进行临床试验。