Multivariate methods for identifying multitask/multimodal brain imaging biomarkers

识别多任务/多模式脑成像生物标志物的多变量方法

基本信息

批准号：
10434066
负责人：
VINCE D CALHOUN
金额：
$ 52.13万
依托单位：
GEORGIA STATE UNIVERSITY
依托单位国家：
美国
项目类别：
财政年份：
2007
资助国家：
美国
起止时间：
2007-04-01 至 2024-06-30
项目状态：
已结题

来源：
https://reporter.nih.gov/project-details/10434066
关键词：
Address Algorithms Area Attention Biological Markers Biology Bipolar Disorder Blood flow Brain Brain Diseases Brain imaging Classification Clinical Communities Complex Data Data Set Databases Decision Making Diagnosis Dictionary Diffusion Magnetic Resonance Imaging Dimensions Disease Documentation Electroencephalography Evaluation Functional Magnetic Resonance Imaging Funding Goals Human Hybrids Image Imaging technology Intuition Joints Link Measurable Measures Mental Depression Mental disorders Methods Modality Modeling Moods Multimodal Imaging Neurobiology Patients Pattern Pharmaceutical Preparations Phase Positioning Attribute Process Psychoses Pythons Reporting Research Research Personnel Rest Sample Size Sampling Schizophrenia Source Structure Symptoms Testing Training Unipolar Depression Validation Work base blind clinical care clinical phenotype data fusion data reduction deep learning design feature extraction gray matter high dimensionality imaging biomarker imaging modality improved large datasets multimodal data multimodality multiple datasets multitask neuropsychiatric disorder next generation novel novel strategies open source open source tool patient subsets potential biomarker repository simulation spatiotemporal tool translational impact user friendly software user-friendly white matter

项目摘要

Project Summary/Abstract The brain is extremely complex as we know, and involves a complicated interplay between functional infor- mation interacting with a structural (but not static) substrate. Brain imaging technology provides a way to sample various aspects of the brain albeit incompletely, providing a rich set of multitask and multimodal information. The field has advanced significantly in its approach to multimodal data, as there are more studies correlating, e.g. functional and structural measures. However, the vast majority of studies still ignore the joint information among two or more modalities or tasks. Such information is critical to consider as each brain imaging modality reports on a different aspect of the brain (e.g. gray matter integrity, blood flow changes, white matter integrity). The field is still striving to understand how to diagnose and treat complex mental illness, such as schizophrenia, bipolar disorder, depression, and others, and ignoring the joint information among tasks and modalities misses a critical, but available, part of the puzzle. Combining multimodal imaging data is not easy since, among other reasons, the combination of multiple data sets consisting of thousands of voxels or timepoints yields a very high dimen- sional problem, requiring appropriate data reduction strategies. In the previous phase of the project we devel- oped advanced approaches to capture high-dimensional relationships among 2 or more modalities. Our work continues to strongly support the benefits of multimodal data fusion to both provide a more complete picture of brain function and structure, but also to improve our ability to study and predict the impact of complex mental illness. In this new phase of the project, we will focus on methods that can fill some existing gaps, such as the ability to bridge spatial/temporal as well as structural/functional connectivity scales. We also propose a novel framework to integrate unimodal and multimodal features called chromatic fusion, which searches for combina- tions of multimodal `notes' which occupy a unique position in a latent (or dictionary) space. The proposed meth- ods will be validated using simulations, hybrid-data, and large N normative imaging data. Our proposed approach will be thoroughly tested using this large data set which includes multiple illnesses that have overlapping symp- toms and which can sometimes be misdiagnosed and treated with the wrong medications for months or years (schizophrenia, bipolar disorder, and unipolar depression). We will provide open source tools and release data throughout the duration of the project via GitHub and the NITRIC repository, hence enabling other investigators to compare their own methods with our own as well as to apply them to a large variety of brain disorders. 37

项目摘要/摘要据我们所知，大脑非常复杂，并且涉及功能性信息之间的复杂相互作用与结构（但不是静态）底物相互作用。脑成像技术提供了一种样品的方法大脑的各个方面尽管并非完全，提供了丰富的多任务和多模式信息。这由于更多的研究相关，例如功能和结构措施。但是，绝大多数研究仍然忽略了共同信息两种或多种方式或任务。这样的信息对于每种大脑成像方式报告至关重要在大脑的不同方面（例如灰质完整性，血流变化，白质完整性）。领域仍在努力了解如何诊断和治疗复杂的精神疾病，例如精神分裂症，双极障碍，抑郁症和其他人，忽略任务和方式之间的共同信息错过了关键，但可用，这是难题的一部分。组合多模式成像数据并不容易，因为除其他原因，由数千个体素或时间点组成的多个数据集的组合产生的尺寸很高需要适当的数据减少策略。在该项目的前阶段，我们开发 OPED先进的方法以捕获两种或更多方式之间的高维关系。我们的工作继续强烈支持多模式数据融合的好处，以提供更完整的图片大脑功能和结构，同时也提高我们学习和预测复杂心理影响的能力疾病。在项目的这个新阶段，我们将重点关注可以填补一些现有空白的方法，例如能够桥接空间/时间和结构/功能连接量表的能力。我们也提出了一本小说整合称为色融合的单形和多模式特征的框架，该特征搜索组合多模式的“音符”，在潜在（或词典）空间中占据了独特的位置。提出的甲基苯丙胺 OD将使用模拟，混合数据和大型N规范成像数据进行验证。我们提出的方法将使用此大数据集对其进行彻底测试，该数据集包括多种疾病，这些疾病重叠了Symp- TOMS有时会被误诊和用错误的药物误诊和治疗（精神分裂症，躁郁症和单极抑郁症）。我们将提供开源工具并发布数据在项目的整个过程中，通过GitHub和Nitric存储库，使其他研究人员能够将自己的方法与我们自己的方法进行比较，并将其应用于各种各样的脑部疾病。 37