Multi-phase CFD modelling of biological tissue flows with applications in liver surgery

计算流体动力学(CFD)已被广泛应用于医学领域，用来模拟大动脉、静脉和心脏中的血液流动，但我们对体液在微观水平上的空间传播的了解还比较有限。该博士项目的目的是通过开发多相CFD模型来研究健康和疾病生物组织中的流动、扩散和渗透性，从而研究这些过程。模型的开发和验证将以电子模拟以及磁共振成像(MRI)和指示剂稀释实验获得的4D动态数据为指导。具体地说，该项目将为几个器官和疾病(如大脑、肌肉、肝脏和癌症)的组织流动产生一套时空模型，提供与测量数据的准确匹配。对于从给定的流场和边界条件生成4D数据的正问题，以及基于测量数据映射流场的反问题，将提供数值解。一组简化的数字组织和器官将被耦合到MRI扫描仪的模型上，为在现实的实验条件下对这些解决方案进行计算机评估提供基本事实。这种全新的分析和解释4D动态MRI数据的方法不仅将显著提高当前1D模型的准确性，还可以揭示营养物质如何输送到健康和患病的生物组织的新见解。此外，这将增加我们对疾病如何影响器官功能和结构的了解，并可能通过更好的诊断和更个性化的治疗计划来及时改善患者的预后。在最后阶段，我们将评估这些方法的应用如何通过更好地预测手术风险来提高肝癌患者的存活率。这将通过开发一种考虑组织流传播空间模式的肝脏手术计划系统来探索临床效用来实现。这个项目涉及来自不同背景(医学成像、物理和土木工程)的学术主管和跨国制药公司拜耳股份公司的一个工业赞助商之间的合作。学生将有一个独特的机会深入了解多相CFD建模和数据同化技术的理论和数值基础，以及它们在医学中的应用-CFD的一个主要增长领域。行业赞助商将在柏林提供一个临时安置的机会，学生将在那里学习如何将研究想法转化为患者的利益。此外，学生将与工业主管合作，将本项目中提出的模型打包成原型软件工具，并利用该工具探索未来商业化的机会。

Computational Fluid Dynamics (CFD) has been used extensively in medicine to model the flow of blood in large arteries and veins and in the heart, but our understanding of the spatial propagation of body fluids on the microscopic level is more limited. The purpose of this PhD project is to enable the study of these processes by developing multi-phase CFD models for flow, diffusion and permeability in healthy and diseased biological tissues. Model development and validation will be guided by in-silico modelling as well as 4D dynamic data obtained by Magnetic Resonance Imaging (MRI) and indicator-dilution experiments.Specifically, the project will produce a set of spatiotemporal models for tissue flow in several organs and diseases (such as brain, muscle, liver and cancer) that provide an accurate fit to measured data. Numerical solutions will be provided for the forward problem of generating 4D data from a given flow fields and boundary conditions, and for the inverse problem of mapping the flow fields based on measured data. A set of simplified digital tissues and organs will be coupled to models of the MRI scanner to provide a ground truth for the in-silico evaluation of these solutions under realistic experimental conditions.This entirely novel approach to analysing and interpreting 4D dynamic MRI data will not only significantly improve upon the accuracy of current 1D models but can also reveal new insights on the way nutrients are delivered to healthy and diseased biological tissue. Furthermore, this will increase our understanding of how disease affects the function and structure of organs, and may in time improve patient outcomes by better diagnosis and more personalised treatment planning. In a final stage we will assess how application of these methods can improve survival of patients with liver cancer through better predictions of surgical risk. This will be achieved by exploring the clinical utility by developing a planning system for liver surgery that accounts for spatial patterns of tissue flow propagation. This project involves a collaboration between academic supervisors from various backgrounds (Medical Imaging, Physics and Civil Engineering), and an industrial sponsor in the multinational pharmaceutical company, Bayer AG. The student will have a unique opportunity to gain in-depth knowledge of the theoretical and numerical basis of multi- phase CFD modelling and data assimilation techniques, but also about their applications in medicine - a major growth area for CFD. The industrial sponsor will provide an opportunity for a temporary placement in Berlin where the student will learn how to translate research ideas into patient benefit. Additionally, the student will, in collaboration with the industrial supervisor, package the models proposed in this project up into a prototype software tool and use this to explore the opportunities for future commercialisation.