Vilcek Institute of Graduate Biomedical Science, New York University School of Medicine; Center for Biomedical Imaging, Radiology, New York University School of Medicine; Center for Advanced Imaging Innovation and Research, Radiology, New York University School of Medicine; Department of Radiology, Weill Cornell Medical College.
Vilcek Institute of Graduate Biomedical Science, New York University School of Medicine; Center for Biomedical Imaging, Radiology, New York University School of Medicine; Center for Advanced Imaging Innovation and Research, Radiology, New York University School of Medicine.
Neuroimage. 2023 Sep;278:120284. doi: 10.1016/j.neuroimage.2023.120284. Epub 2023 Jul 26.
In Dynamic contrast-enhanced MRI (DCE-MRI), Arterial Input Function (AIF) has been shown to be a significant contributor to uncertainty in the estimation of kinetic parameters. This study is to assess the feasibility of using a deep learning network to estimate local Capillary Input Function (CIF) to estimate blood-brain barrier (BBB) permeability, while reducing the required scan time.
A total of 13 healthy subjects (younger (<40 y/o): 8, older (> 67 y/o): 5) were recruited and underwent 25-min DCE-MRI scans. The 25 min data were retrospectively truncated to 10 min to simulate a reduced scan time of 10 min. A deep learning network was trained to predict the CIF using simulated tissue contrast dynamics with two vascular transport models. The BBB permeability (PS) was measured using 3 methods: (i) C-25min, using DCE-MRI data of 25 min with individually sampled AIF (C); (ii) C-10min, using truncated 10min data with AIF (C); and (iii) C-10min, using truncated 10 min data with CIF (C). The PS estimates from the C-25min method were used as reference standard values to assess the accuracy of the C-10min and C-10min methods in estimating the PS values.
When compared to the reference method(C-25min), the C-10min and C-10min methods resulted in an overestimation of PS by 217 ± 241 % and 48.0 ± 30.2 %, respectively. The Bland Altman analysis showed that the mean difference from the reference was 8.85 ± 1.78 (x10 min) with the C-10min, while it was reduced to 1.63 ± 2.25 (x10 min) with the C-10min, resulting in an average reduction of 81%. The limits of agreement also reduced by up to 39.2% with the C-10min. We found a 75% increase of BBB permeability in the gray matter and a 35% increase in the white matter, when comparing the older group to the younger group.
We demonstrated the feasibility of estimating the capillary-level input functions using a deep learning network. We also showed that this method can be used to estimate subtle age-related changes in BBB permeability with reduced scan time, without compromising accuracy. Moreover, the trained deep learning network can automatically select CIF, reducing the potential uncertainty resulting from manual user-intervention.
在动态对比增强磁共振成像（DCE-MRI）中，动脉输入函数（AIF）已被证明是对动力学参数估计不确定性的重要贡献因素。本研究旨在评估使用深度学习网络估计局部毛细血管输入函数（CIF）以估计血脑屏障（BBB）通透性的可行性，同时减少所需的扫描时间。
共招募了 13 名健康受试者（年轻组（<40 岁）：8 名，老年组（>67 岁）：5 名），并进行了 25 分钟的 DCE-MRI 扫描。回顾性地将 25 分钟的数据截断为 10 分钟，以模拟减少的 10 分钟扫描时间。使用两种血管转运模型，使用模拟组织对比动力学训练深度学习网络来预测 CIF。使用 3 种方法测量 BBB 通透性（PS）：（i）C-25min，使用 25 分钟 DCE-MRI 数据和个体采样的 AIF（C）；（ii）C-10min，使用截断的 10 分钟数据和 AIF（C）；和（iii）C-10min，使用截断的 10 分钟数据和 CIF（C）。将 C-25min 方法的 PS 估计值用作参考标准值，以评估 C-10min 和 C-10min 方法在估计 PS 值方面的准确性。
与参考方法（C-25min）相比，C-10min 和 C-10min 方法导致 PS 分别高估了 217±241%和 48.0±30.2%。Bland-Altman 分析显示，与参考值的平均差值为 8.85±1.78（x10min），而 C-10min 为 1.63±2.25（x10min），平均降低了 81%。与 C-10min 相比，一致性界限也降低了高达 39.2%。与年轻组相比，我们发现老年组的脑灰质 BBB 通透性增加了 75%，白质增加了 35%。
我们证明了使用深度学习网络估计毛细血管水平输入功能的可行性。我们还表明，该方法可用于在不影响准确性的情况下，估计扫描时间减少时 BBB 通透性的细微年龄相关变化。此外，经过训练的深度学习网络可以自动选择 CIF，减少手动用户干预带来的潜在不确定性。
Zotero 插件
