Assessing the health of ageing blood vessels in the brain using fMRI: experimental design, modelling and analysis

在人口老龄化过程中，与控制脑血流有关的神经系统问题日益增多，是导致阿尔茨海默病和其他形式的痴呆症等与衰老有关的病症的一个因素。如果大脑小动脉周围的平滑肌功能不正常，就可能供血过少，导致神经功能出现问题。目前，我们正在开发实用的功能性磁共振成像（fMRI）工具来评估大脑血管的健康状况。这个博士项目的目标是开发新的功能磁共振成像技术和计算软件，以改善脑血管健康的测量。这些临床技术的效用将在合适的患者群体中进行评估。首先，学生将在对大脑血管施加压力的二氧化碳挑战中获取MR数据。由于这种类型的挑战通常会在软件中与头部运动混淆，因此运动跟踪相机系统将作为黄金标准串联使用。将开发运动混淆的量化和数据特征提取方法，以提高从这些数据中获得的船舶健康措施的可靠性。第二种方法将利用现有的大规模成像数据库，通过开发新的信号处理技术来获得新的血管健康测量方法。当前的噪声校正方法丢弃了我们感兴趣的信息，因此这将需要对所采用的逻辑进行反转。在博士课程的这一部分，学生将开发处理大数据的算法。最后，通过设计更舒适、对患者友好的磁共振兼容装置，同时改进相关的处理管道，将寻求对生理挑战管理方式的实际改进。将开发一种下体负压室，以最大限度地减少扫描仪中的运动。目的是在中等规模的患者队列中增加相关血管健康测量的可重复性。

In an ageing population, neurological problems related to control of brain blood flow are increasing and are a contributing factor in conditions associated with ageing, such as Alzheimer's disease and other forms of dementia. If the smooth muscle around small arteries in the brain is not functioning well, too little blood may be supplied, causing problems for neurological function. Currently, we are developing pragmatic functional magnetic resonance imaging (fMRI) tools to assess the health of the brain's blood vessels. The objective of this PhD project is to develop novel fMRI imaging techniques and computational software to improve measures of brain vessel health. The utility of these clinical techniques will then be assessed in a suitable patient population. Initially, the student will acquire MR data during carbon dioxide challenges that stress the brain's blood vessels. Since this type of challenge can often be confused in software as head motion, a motion tracking camera system will be used in tandem as a gold standard. Quantification of motion confounds and data feature extraction methods will be developed to increase the reliability of vessel health measures obtained from this data. The second approach will take advantage of existing large scale imaging databases to derive new measure of vessel health by developing novel signal processing techniques. Current noise correction methods discard the information we are interested in and so this will require an inversion of the employed logic. In this section of the PhD, the student will develop algorithms to deal with large data. Finally, practical improvements to the way physiological challenges are administered will be sought by engineering more comfortable, patient- friendly MR compatible setups and by simultaneously improving related processing pipelines. A lower body negative pressure chamber that minimises motion in the scanner will be developed. The objective is to increase the repeatability of related vessel health measures in a medium sized patient cohort.