Three-dimensional analysis of hemodynamic wall parameters in the carotid bifurcation using computer-augmented 4D flow MRI

颈内动脉（ICA）狭窄是中风的主要原因。心血管危险因素、斑块组成、颈动脉分叉几何形状、壁剪切应力（WSS）和斑块表面压力（PSP）被认为是ICA动脉粥样硬化进展和破裂的危险因素。MRI非常适合表征斑块形态，4D血流MRI可以测量体内3D血流，并通过专用软件计算WSS和PSP。然而，4D流动MRI的缺点是其有限的空间和时间分辨率，阻碍了壁面参数的详细和准确计算。这是晚期腔狭窄和血流加速的ICA狭窄的一个特殊问题。此外，后处理软件还不能充分利用所获得的三维流动信息，因此可视化和量化一直局限于二维水平。最后，精确计算作用于斑块表面的WSS和PSP对预测斑块进展、破裂和中风非常有希望。然而，这些计算具有挑战性，因此尚未与体内ICA狭窄的斑块组成相关。为了克服这些限制，我们计划使用数据同化方法将4D流动MRI与计算流体动力学相结合，以提高测量流场的质量（计算机增强4D流动MRI）。首先，从先前的患者研究中获得的4D血流数据将用于定制现有的数据同化框架，该框架使用高阶Navier-Stokes求解器来求解颈动脉分叉的流场和斑块表面的WSS和压力预测，并优化数据的3D表示。之后，我们将招募至少100例≥50% ICA狭窄且斑块进展风险高的患者，并对他们进行24个月的研究。一些患者会出现无症状的狭窄进展和/或出现症状并进行手术。该前瞻性患者数据将通过数据同化框架得到增强。由此产生的计算机增强MRI数据将用于识别斑块稳定性或进展的独立预测因素，通过基线检查与研究结束时或手术前的随访进行比较。我们假设低绝对剪应力或高振荡剪应力或高PSP是斑块进展、破裂和脑血管事件的独立预测因子。总之，我们的目标是通过四维血流MRI获得并通过高阶计算流体动力学的数据同化过程增强对≥50% ICA狭窄的3D颈动脉血流动力学进行最佳测量和解释。我们的目标是为颈动脉斑块患者开发一种改进的诊断工具。这可能有助于优化个体治疗，并最终通过确定新的危险因素（如WSS和PSP）来预防脑血管事件，这些因素会破坏动脉粥样硬化的稳定性。

Internal carotid artery (ICA) stenoses are a major source of stroke. Cardiovascular risk factors, plaque composition, carotid bifurcation geometry, wall shear stress (WSS), and plaque surface pressure (PSP) are considered as risk factors for the progression and rupture of ICA atheroma. MRI is ideally suited to characterize plaque morphology, and 4D flow MRI allows to measure 3D blood flow in vivo, and to calculate WSS and PSP by employing dedicated software. However, the drawback of 4D flow MRI is its limited spatial and temporal resolution that hampers detailed and accurate calculation of wall parameters. This is a particular problem in ICA stenoses with advanced lumen narrowing and blood flow acceleration. In addition, post-processing software has not yet been able to fully use the acquired 3D flow information wherefore visualization and quantification has been restricted to a 2D level. Finally, the precise calculation of WSS and PSP acting on the plaque surface are highly promising to allow prediction of plaque progression, rupture and stroke. However, these calculations are challenging and thus have not yet been correlated with plaque composition in ICA stenosis in vivo. To overcome these limitations, we plan to use a data assimilation approach to combine 4D flow MRI with computational fluid dynamics in order to enhance the quality of the measured flow field (computer-augmented 4D flow MRI). At first, available 4D flow data from a previous study in patients will be used to customize an existing data assimilation framework with a high-order Navier-Stokes solver for flow fields in the carotid bifurcation and the WSS and pressure prediction on the plaque surface and optimal 3D presentation of data. Afterwards, we will recruit at least 100 patients with ≥50% ICA stenosis with a high risk of plaque progression and study them over 24 months. Some patients will show asymptomatic progression of stenosis and/or become symptomatic and undergo surgery. This prospective patient data will be enhanced with the data assimilation framework. The resulting computer-augmented MRI data will be used to identify independent predictors of plaque stability or progression using baseline examinations in comparison with follow-up at the end of the study or prior to surgery. We hypothesize that low absolute or high oscillating shear stress or high PSP are independent predictors of plaque progression, rupture, and cerebrovascular events.In conclusion, we aim to optimally measure and interpret 3D carotid hemodynamics, obtained by 4D flow MRI and enhanced through a data assimilation process with high-order computational fluid dynamics in ≥50% ICA stenosis. It is our goal to develop an improved diagnostic tool for patients with carotid artery plaque. This may help to optimize individual treatment and ultimately prevent cerebrovascular events by determining new risk factors such as WSS and PSP, which destabilize the atheroma.