Computational biophysical modelling for the optimisation of cardiac magnetic resonance perfusion imaging protocols

心肌梗死（MI）是一种严重的疾病，当血液流向心脏的某一部分受到限制或停止时，就会发生心肌损伤。由此造成的心脏损伤包括心肌细胞的纤维化和坏死，随着时间的推移，这些损伤会影响心脏有效泵血的能力，最终导致部分（但不是全部）患者心力衰竭。因此，有必要了解造成这种损害的过程，从而了解观察到的不同临床结果。这就为制定针对患者的治疗方案提供了机会，从而限制这些负面结果。该项目将采用MI期间发生的各种细胞过程的数学建模和计算模拟，目的是提供度量或计算工具的原理证明，以便在治疗疑似MI患者时为临床医生提供诊断支持。建议使用细胞过程的常微分方程模型，并在适当的情况下使用心脏力学的偏微分方程模型。这将由布里斯托尔大学转化生物医学研究中心的合作者从一系列猪实验中收集的数据来参数化。这些数据将包括血液生物标志物和定量成像数据，由于数据收集将在心肌梗死期间和随后的时期进行，因此这些数据将为验证这些模型的预测能力提供充足的手段。该项目的一个更广泛的目标是创建模型，最初可以使用更大的数据集进行验证，并最终使用临床数据进行测试。

Myocardial infarction (MI) is a severe condition that occurs when blood flow is restricted or stopped to a part of the heart, causing damage to the heart muscle. Resulting injury to the heart can include fibrosis and necrosis of myocytes, which overtime can affect the ability of the heart to pump efficiently, ultimately leading to heart failure in some, but not all, patients. Thus, there is a need to understand the processes that are responsible for causing this damage, and to consequently understand the differing clinical outcomes that are observed. This then provides an opportunity for deriving patient-specific treatments that will limit these negative outcomes.This project will employ mathematical modelling and computational simulation of the various cellular processes that occur during MI, with the aims of providing a proof of principle of a metric or computational tool that will provide diagnostic support to clinicians when treating patients with suspected MI.It is proposed to use ordinary differential equation models of the cellular processes, and where appropriate partial differential equation models of cardiac mechanics, which will be parameterised by data collected from a series of porcine experiments performed by collaborators at the Translational Biomedical Research Centre, University of Bristol. This data will comprise both blood biomarker and quantitative imaging data, and as the data collection will occur both during the MI and the subsequent period, the data will provide ample means to validate these models' predictive capabilities. A broader aim of this project is to create models that can be validated using larger data sets initially, and ultimately be taken forward to test using clinical data.