权益分类	功能权益	普通用户	{{item.name}}会员
{{category.name}}	{{benefitItem.name}}

Multimodal MR-PET Machine Learning Approaches for Primary Prostate Cancer Characterization

用于原发性前列腺癌表征的多模态 MR-PET 机器学习方法

基本信息

批准号：
10557135
负责人：
Ciprian Catana
金额：
$ 23.01万
依托单位：
MASSACHUSETTS GENERAL HOSPITAL
依托单位国家：
美国
项目类别：
财政年份：
2018
资助国家：
美国
起止时间：
2018-02-21 至 2025-01-31
项目状态：
未结题

来源：
https://reporter.nih.gov/project-details/10557135
关键词：
3-Dimensional Acceleration Address Affect Algorithms Anatomy Biological Biopsy Cancer Death Rates Cancer Patient Classification Computer software Data Data Set Devices Diagnosis Diagnostic Discrimination Disease Early Diagnosis Early treatment FOLH1 gene Guidelines Hand Histopathology Hybrids Image Individual Interobserver Variability Kinetics Lesion Life Expectancy Ligands Machine Learning Magnetic Resonance Imaging Malignant Neoplasms Malignant neoplasm of prostate Maps Measurement Meta-Analysis Metabolic Methodology Methods Modality Modeling Morphology PSA level Patient Selection Patients Pelvis Performance Phenotype Physiological Positron-Emission Tomography Prostate Quality of life Radical Prostatectomy Reproducibility Risk Scanning Source Testing Time Translations Work anticancer research attenuation bone imaging cancer classification cancer imaging clinical application clinically relevant clinically significant data acquisition deep learning diagnostic accuracy human subject imaging modality improved industry partner machine learning model men multimodal data multimodality new technology non-invasive imaging radiologist radiomics radiotracer tool transmission process tumor

项目摘要

Project Summary Prostate cancer (PCa) is the most diagnosed form of non-cutaneous cancer in US men. The selection of patients who require immediate treatment from those suitable for active surveillance currently relies on non- specific and inaccurate measurements. A method that allows clinicians to more confidently discriminate clinically relevant from non-life-threatening tumors is needed to improve patient management. Multiparametric magnetic resonance imaging (mpMRI) is the preferred non-invasive imaging modality for characterizing primary PCa. However, its accuracy for detecting clinically significant PCa is variable. We propose to address this limitation by combining mpMRI with positron emission tomography (PET) with a PCa-specific radiotracer and using advanced multimodal machine learning models (i.e. radiomics and deep learning) to characterize tumor aggressiveness based on the imaging data. Recently, scanners capable of simultaneous PET and MR data acquisition in human subjects have become commercially available. An integrated MR-PET scanner is the ideal tool for comparing MR and PET derived image features to identify those that provide complementary information and build a hybrid PET-mpMRI model that most accurately identifies clinically significant tumors. While this novel technology allows the acquisition of perfectly coregistered complementary anatomical, functional and metabolic data in a single imaging session, a new challenge needs to first be addressed to obtain quantitatively accurate PET data. In an integrated MR-PET scanner, the information needed for PET attenuation correction (AC) has to be derived from the MR data and the methods currently available for this task are inadequate for advanced quantitative studies. We have formed an academic-industrial partnership to accelerate the translation of multimodal MR-PET machine learning approaches into PCa research and clinical applications by addressing the AC challenge and validating machine learning models for detecting clinically significant disease against gold standard histopathology in patients undergoing radical prostatectomy. Specifically, we will: (1) Develop and validate an MR-based approach for obtaining quantitatively accurate PET data. We hypothesize that attenuation maps as accurate as those obtained using a 511 keV transmission source – the true gold standard for PET AC – will be obtained; (2) Identify the multimodal radiomics model that most accurately predicts PCa aggressiveness. We hypothesize that the diagnostic accuracy of this approach will be superior to that offered by the stand-alone modalities; (3) Evaluate radiomics and deep learning approaches for predicting pPCa aggressiveness. We hypothesize that machine learning approaches will achieve a higher predictive accuracy when applied to data acquired simultaneously than sequentially.

项目概要前列腺癌 (PCa) 是美国男性中诊断最多的非皮肤癌。的选择需要立即接受适合主动监测的治疗的患者目前依赖于非具体且不准确的测量。一种让临床医生能够更加自信地进行区分的方法需要与非危及生命的肿瘤具有临床相关性来改善患者管理。多参数磁共振成像 (mpMRI) 是表征特征的首选非侵入性成像方式原发性前列腺癌。然而，其检测具有临床意义的 PCa 的准确性是可变的。我们建议解决通过将 mpMRI 与正电子发射断层扫描 (PET) 和 PCa 特异性放射性示踪剂相结合来克服这一限制并使用先进的多模态机器学习模型（即放射组学和深度学习）来表征基于成像数据的肿瘤侵袭性。最近，能够同时进行 PET 和 MR 扫描的扫描仪人类受试者的数据采集已经商业化。集成 MR-PET 扫描仪是比较 MR 和 PET 衍生图像特征以识别那些提供互补的图像特征的理想工具信息并构建混合 PET-mpMRI 模型，该模型可以最准确地识别具有临床意义的肿瘤。虽然这项新技术可以获取完美配准的互补解剖结构，在一次成像过程中获得功能和代谢数据，首先需要解决一个新的挑战获得定量准确的 PET 数据。在集成 MR-PET 扫描仪中，PET 所需的信息衰减校正 (AC) 必须从 MR 数据和当前可用的方法中得出任务不足以进行高级定量研究。我们建立了学术与工业合作伙伴关系加速将多模态 MR-PET 机器学习方法转化为 PCa 研究和临床通过解决 AC 挑战并验证用于临床检测的机器学习模型来应用在接受根治性前列腺切除术的患者中，存在与金标准组织病理学不符的显着疾病。具体来说，我们将： (1) 开发并验证基于 MR 的方法，以获得定量准确的 PET 数据。我们假设衰减图与使用 511 keV 传输获得的衰减图一样准确来源——PET AC 的真正黄金标准——将获得； (2) 确定多模态放射组学模型最准确地预测 PCa 侵袭性。我们假设这种方法的诊断准确性将优于独立模式所提供的服务； (3) 评估放射组学和深度学习预测 pPCa 侵袭性的方法。我们假设机器学习方法将当应用于同时获取的数据时，比顺序获取的数据获得更高的预测精度。

项目成果

期刊论文数量（8）

专著数量（0）

科研奖励数量（0）

会议论文数量（0）

专利数量（0）

Attenuation correction for human PET/MRI studies.

DOI：
10.1088/1361-6560/abb0f8
发表时间：
2020-12-02
期刊：
Physics in medicine and biology
影响因子：
3.5
作者：
Catana C
通讯作者：
Catana C

Detection and Characterization of Thrombosis in Humans Using Fibrin-Targeted Positron Emission Tomography and Magnetic Resonance.

DOI：
10.1016/j.jcmg.2021.08.009
发表时间：
2022-03
期刊：
JACC. Cardiovascular imaging
影响因子：
0
作者：
Izquierdo-Garcia D;Désogère P;Philip AL;Mekkaoui C;Weiner RB;Catalano OA;Iris Chen YC;DeFaria Yeh D;Mansour M;Catana C;Caravan P;Sosnovik DE
通讯作者：
Sosnovik DE

Evaluation of Deep Learning-Based Approaches to Segment Bowel Air Pockets and Generate Pelvic Attenuation Maps from CAIPIRINHA-Accelerated Dixon MR Images.

基于深度学习的方法的评估，用于分割肠道气囊并从 CAIPIRINHA 加速的 Dixon MR 图像生成骨盆衰减图。

DOI：
10.2967/jnumed.120.261032
发表时间：
2022
期刊：
Journal of nuclear medicine : official publication, Society of Nuclear Medicine
影响因子：
0
作者：
Sari,Hasan;Reaungamornrat,Ja;Catalano,OnofrioA;Vera-Olmos,Javier;Izquierdo-Garcia,David;Morales,ManuelA;Torrado-Carvajal,Angel;Ng,ThomasSC;Malpica,Norberto;Kamen,Ali;Catana,Ciprian
通讯作者：
Catana,Ciprian

A Path to Qualification of PET/MRI Scanners for Multicenter Brain Imaging Studies: Evaluation of MRI-Based Attenuation Correction Methods Using a Patient Phantom.

PET/MRI扫描仪进行多中心脑成像研究的资格途径：使用患者幻影评估基于MRI的衰减校正方法。

DOI：
10.2967/jnumed.120.261881
发表时间：
2022-04
期刊：
Journal of nuclear medicine : official publication, Society of Nuclear Medicine
影响因子：
0
作者：
Catana C;Laforest R;An H;Boada F;Cao T;Faul D;Jakoby B;Jansen FP;Kemp BJ;Kinahan PE;Larson P;Levine MA;Maniawski P;Mawlawi O;McConathy JE;McMillan AB;Price JC;Rajagopal A;Sunderland J;Veit-Haibach P;Wangerin KA;Ying C;Hope TA
通讯作者：
Hope TA

Multimodal image synthesis based on disentanglement representations of anatomical and modality specific features, learned using uncooperative relativistic GAN.