Improving Arterial Spin Labeling by Using Deep Learning
Improving Arterial Spin Labeling by Using Deep Learning
复制标题
DOI：
10.1148/radiol.2017171154
复制
发表时间：
2018-05-01
期刊：
影响因子：
19.7
通讯作者：
Park, Sung-Hong
中科院分区：
文献类型：
--
作者：
Kim, Ki Hwan;Choi, Seung Hong;Park, Sung-Hong
目的：开发一种深度学习算法，通过使用较少数量的减法图像来生成具有更高准确性和鲁棒性的动脉自旋标记（ASL）灌注图像。材料和方法：对于通过成对减法生成 ASL 图像，我们使用卷积神经网络（CNN）作为深度学习算法。地面真实灌注图像是通过对六或七对减影图像进行平均而生成的，这些图像是通过（a）来自七名健康受试者的传统伪连续动脉自旋标记或（b）来自 114 名患有各种疾病的患者的 Hadamard 编码的伪连续 ASL 获得的。 CNN 被训练为从少量（两个或三个）减影图像生成灌注图像，并通过交叉验证进行评估。来自患者数据集的 CNN 也在 26 个独立的中风数据集上进行了测试。通过配对 t 检验和/或 Wilcoxon 符号秩检验，将 CNN 与传统平均方法的均方误差和放射学评分进行比较。结果：与健康受试者和患者的交叉验证以及与经历过中风的患者的单独测试相比，均方误差比传统平均方法低约 40% (P < .001)。中风区域的感兴趣区域分析表明，CNN 的脑血流图（平均值 +/- 标准偏差，19.7 mL 每 100 g/min +/- 9.7）的均方误差比使用传统平均方法确定的脑血流量图 (43.2 +/- 29.8) 更小 (P < .001)。放射学评分表明，CNN 比传统平均方法更好地抑制了噪声、运动和/或分割伪影 (P < .001)。 结论：与从成对减影图像生成 ASL 图像的传统平均方法相比，CNN 提供了卓越的灌注图像质量和更准确的灌注测量。 (C) 北美放射学会，2017
Purpose: To develop a deep learning algorithm that generates arterial spin labeling (ASL) perfusion images with higher accuracy and robustness by using a smaller number of subtraction images.Materials and Methods: For ASL image generation from pair-wise subtraction, we used a convolutional neural network (CNN) as a deep learning algorithm. The ground truth perfusion images were generated by averaging six or seven pairwise subtraction images acquired with (a) conventional pseudocontinuous arterial spin labeling from seven healthy subjects or (b) Hadamard-encoded pseudocontinuous ASL from 114 patients with various diseases. CNNs were trained to generate perfusion images from a smaller number (two or three) of subtraction images and evaluated by means of cross-validation. CNNs from the patient data sets were also tested on 26 separate stroke data sets. CNNs were compared with the conventional averaging method in terms of mean square error and radiologic score by using a paired t test and/or Wilcoxon signed-rank test.Results: Mean square errors were approximately 40% lower than those of the conventional averaging method for the cross-validation with the healthy subjects and patients and the separate test with the patients who had experienced a stroke (P < .001). Region-of-interest analysis in stroke regions showed that cerebral blood flow maps from CNN (mean +/- standard deviation, 19.7 mL per 100 g/min +/- 9.7) had smaller mean square errors than those determined with the conventional averaging method (43.2 +/- 29.8) (P < .001). Radiologic scoring demonstrated that CNNs suppressed noise and motion and/or segmentation artifacts better than the conventional averaging method did (P < .001).Conclusion: CNNs provided superior perfusion image quality and more accurate perfusion measurement compared with those of the conventional averaging method for generation of ASL images from pair-wise subtraction images. (C)RSNA, 2017