Development of flow and vascular quantification software for the assessment of MR

描述(申请人提供)：对多发性硬化症、创伤性脑损伤、中风和痴呆症等疾病的临床环境中产生的神经血管成像数据的全面定量分析的需求大幅增加。我们在这个项目中的目标是设计和开发先进的图像处理软件，可以快速准确地分析这些数据。为了实现这一目标，我们提出了一系列新的算法来处理来自以下MR成像序列的数据：用于评估血管解剖的时间分辨3D对比增强MR血管成像(CE-MRA)、用于评估血管血流动力学的时间分辨2D相位对比流动成像(PC-MRI)、用于定量脑内铁沉积的磁化率加权成像(SWI)以及用于检测白质高信号(WMH)和病变的液体衰减反转恢复(FLAIR)成像。为了解决这些问题，将设计和实施各种工具，包括：组织相似性图谱和活动形状模型，用于在CE-MRA和PC-MRI图像中分割血管系统；自动分割基底节和丘脑的组织，以实现双感兴趣区；利用SWI进行铁定量分析；以及最后，结合模糊C均值、形状因素分析、致密性和分数各向异性的自适应方法来量化病变和WMH。为了利用不同成像序列的优势，将使用联合配准算法来改进CE-MRA和PC-MRI以及3D T1加权成像和SWI之间的血管分割。完成这个项目后，我们预计处理效率将提高数倍，精确度将显著提高。由此产生的软件不仅有助于我们公司的发展，还将改善神经血管疾病的诊断和治疗。

DESCRIPTION (provided by applicant): There has been a huge increase in demand for comprehensive quantitative analysis of neurovascular imaging data produced in the clinical setting for diseases such as multiple sclerosis, traumatic brain injury, stroke and dementia. Our objective in this project is to design and develop advanced image processing software that can rapidly and accurately analyze such data. To achieve this objective, we propose a range of novel algorithms to process data from the following MR imaging sequences widely used in the aforementioned applications: time resolved 3D contrast enhanced MR angiography (CE-MRA) for the assessment of vascular anatomy, time resolved 2D phase contrast flow imaging (PC-MRI) for the evaluation of vascular hemodynamics, susceptibility weighted imaging (SWI) for quantifying iron deposition in the brain, and fluid attenuated inversion recovery (FLAIR) imaging for the detection of white matter hyperintensities (WMH) and lesions. A variety of tools will be designed and implemented to tackle these problems including: tissue similarity mapping and active shape models to segment the vasculature in both CE-MRA and PC-MRI images; automatic tissue segmentation in the basal ganglia and thalamus for a two-region of interest analysis for iron quantification with SWI; and finally adaptive approaches incorporating fuzzy C-means, shape factor analysis, compactness and fractional anisotropy to quantify lesions and WMHs. To exploit the advantages provided by different imaging sequences, co-registration algorithms will be used to improve segmentation of vessels between CE-MRA and PC-MRI, and between 3D T1 weighted imaging and SWI. Upon finishing this project, we expect a multi-fold increase in processing efficiency and a significant increase in accuracy will be achieved. The resulting software will not only help the growth of our company, but also improve the diagnosis and treatment of neurovascular diseases.