Artificial intelligence velocimetry reveals in vivo flow rates, pressure gradients, and shear stresses in murine perivascular flows.
Artificial intelligence velocimetry reveals in vivo flow rates, pressure gradients, and shear stresses in murine perivascular flows.
复制标题
DOI：
10.1073/pnas.2217744120
复制
发表时间：
2023-04-04
影响因子：
11.1
通讯作者：
Kelley, Douglas H.
中科院分区：
文献类型：
--
作者：
Boster, Kimberly A. S.;Cai, Shengze;Ladron-de-Guevara, Antonio;Sun, Jiatong;Zheng, Xiaoning;Du, Ting;Thomas, John H.;Nedergaard, Maiken;Karniadakis, George Em;Kelley, Douglas H.
阿尔茨海默氏症和小血管病等疾病与脑血管周围的血管周围空间的流动变化有关，这些空间在脑组织周围输送水状液体。了解系统的功能、故障和潜在的恢复取决于对流速、压力和剪切应力的高保真体内量化，而这在以前是无法实现的。我们证明，人工智能测速（AIV）将稀疏二维（2D）体内速度测量与物理信息神经网络相结合，可以准确推断高分辨率压力和剪切应力。 AIV 还可以推断高分辨率三维 (3D) 速度，从而高精度地量化体积流量和阻力。其独特的功能使 AIV 成为了解脑液流动、改善临床结果的关键工具。量化脑脊液 (CSF) 的流量对于了解脑废物清除和营养输送以及中风等病理状况下的水肿至关重要。然而，现有的体内技术仅限于软脑膜血管周围空间（PVS）的稀疏速度测量或全脑成像的低分辨率测量。此外，PVS 中的体积流量、压力和剪切应力变化基本上不可能在体内测量。在这里，我们展示了人工智能测速（AIV）可以将稀疏速度测量与物理信息神经网络相结合，以量化 PVS 中的 CSF 流量。借助 AIV，我们可以推断三维 (3D)、高分辨率速度、压力和剪切应力。验证来自使用 70% PTV 测量值进行的训练，并证明与其余 30% 的测量结果非常一致。对 AIV 输入的敏感性分析表明，由于体内成像固有的 PVS 边界位置的不确定性，AIV 推断量的不确定性小于 30%，神经网络初始化的不确定性小于 1%。在 N = 4 只野生型小鼠的 PVS 中，我们发现平均流速为 16.33 ± 11.09 µm/s，体积流速为 2.22 ± 1.983 × 103 µm3/s，轴向压力梯度 (− 2.75 ± 2.01)×10−4 Pa/µm (−2.07 ± 1.51 mmHg/m) 和壁剪切应力 (3.00 ± 1.51 mmHg/m) 1.45)×10−3 Pa（均为平均值±SE）。压力梯度、流速和阻力与之前的预测一致。 AIV 可以非常详细地推断体内 PVS 流量，这将改善流体动力学模型，并有可能阐明脑脊液流量如何随衰老、阿尔茨海默病和小血管疾病而变化。
Diseases such as Alzheimer’s and small vessel disease are linked to alterations of flow in the perivascular spaces that surround cerebral blood vessels and transport water-like fluids around brain tissue. Understanding the function, failure, and potential rehabilitation of the system depends on high-fidelity, in vivo quantification of flow rates, pressure, and shear stress, which have previously been unavailable. We show that artificial intelligence velocimetry (AIV), which integrates sparse two-dimensional (2D) in vivo velocity measurements with physics-informed neural networks, can accurately infer high-resolution pressure and shear stresses. AIV can also infer high-resolution three-dimensional (3D) velocities, thereby quantifying volume flow rates and resistances with high accuracy. Its unique capabilities make AIV a key tool for understanding brain fluid flow, toward improved clinical outcomes. Quantifying the flow of cerebrospinal fluid (CSF) is crucial for understanding brain waste clearance and nutrient delivery, as well as edema in pathological conditions such as stroke. However, existing in vivo techniques are limited to sparse velocity measurements in pial perivascular spaces (PVSs) or low-resolution measurements from brain-wide imaging. Additionally, volume flow rate, pressure, and shear stress variation in PVSs are essentially impossible to measure in vivo. Here, we show that artificial intelligence velocimetry (AIV) can integrate sparse velocity measurements with physics-informed neural networks to quantify CSF flow in PVSs. With AIV, we infer three-dimensional (3D), high-resolution velocity, pressure, and shear stress. Validation comes from training with 70% of PTV measurements and demonstrating close agreement with the remaining 30%. A sensitivity analysis on the AIV inputs shows that the uncertainty in AIV inferred quantities due to uncertainties in the PVS boundary locations inherent to in vivo imaging is less than 30%, and the uncertainty from the neural net initialization is less than 1%. In PVSs of N = 4 wild-type mice we find mean flow speed 16.33 ± 11.09 µm/s, volume flow rate 2.22 ± 1.983 × 103 µm3/s, axial pressure gradient ( − 2.75 ± 2.01)×10−4 Pa/µm (−2.07 ± 1.51 mmHg/m), and wall shear stress (3.00 ± 1.45)×10−3 Pa (all mean ± SE). Pressure gradients, flow rates, and resistances agree with prior predictions. AIV infers in vivo PVS flows in remarkable detail, which will improve fluid dynamic models and potentially clarify how CSF flow changes with aging, Alzheimer’s disease, and small vessel disease.
登录
查看更多内容
DOI：
10.1016/j.pathophys.2009.09.002
发表时间：
2010-09
期刊：
Pathophysiology : the official journal of the International Society for Pathophysiology
影响因子：
--
作者：
Gashev AA;Zawieja DC
通讯作者：
Zawieja DC
影响因子：
3.7
作者：
Daversin-Catty C;Vinje V;Mardal KA;Rognes ME
通讯作者：
Rognes ME
DOI：
10.1177/0271678x17737984
发表时间：
2018-04
期刊：
Journal of cerebral blood flow and metabolism : official journal of the International Society of Cerebral Blood Flow and Metabolism
影响因子：
--
作者：
Bedussi B;Almasian M;de Vos J;VanBavel E;Bakker EN
通讯作者：
Bakker EN
影响因子：
5.8
作者：
Bohr, Tomas;Hjorth, Poul G.;Holst, Sebastian C.;Hrabetova, Sabina;Kiviniemi, Vesa;Lilius, Tuomas;Lundgaard, Iben;Mardal, Kent-Andre;Martens, Erik A.;Mori, Yuki;Nagerl, U. Valentin;Nicholson, Charles;Tannenbaum, Allen;Thomas, John H.;Tithof, Jeffrey;Benveniste, Helene;Iliff, Jeffrey J.;Kelley, Douglas H.;Nedergaard, Maiken
通讯作者：
Nedergaard, Maiken
DOI：
10.1093/brain/awaa443
发表时间：
2021-04-12
期刊：
Brain : a journal of neurology
影响因子：
--
作者：
Eide, Per Kristian;Vinje, Vegard;Ringstad, Geir
通讯作者：
Ringstad, Geir