Deep-learning-based direct inversion for material decomposition.
Deep-learning-based direct inversion for material decomposition.
复制标题
DOI：
10.1002/mp.14523
复制
发表时间：
2020-12
期刊：
影响因子：
3.8
通讯作者：
Leng S
中科院分区：
文献类型：
--
作者：
Gong H;Tao S;Rajendran K;Zhou W;McCollough CH;Leng S
开发一种卷积神经网络（CNN），可以直接从多能量CT图像中估计材料密度分布，而无需进行常规的材料分解。提出的CNN（表示为Incept-net）遵循编码器-解码器网络的一般框架，假设局部图像信息足以建模多能量CT的非线性物理过程。Incept-net使用定制的损失函数实现，包括内部设计的图像梯度相关（IGC）正则化器，以改善边缘保留。该网络由两种类型的定制多分支模块组成，利用多尺度特征表示来提高对局部图像噪声和伪影的鲁棒性。使用具有两个能量阈值和多个辐射剂量水平的研究光子计数探测器（PCD）CT扫描具有不同密度的不同材料（羟基磷灰石（HA）、碘、血碘混合物和脂肪）的骨。该网络仅使用体模图像块进行训练，并使用全视场体模和体内猪图像的不同配置进行测试。此外，插入物材料的标称质量密度用作CNN训练中的标签，这可能提供隐含的质量守恒约束。Incept网络的性能进行了评估，在图像噪声，细节保留，和定量准确性。它的性能也进行了比较，常见的材料分解算法，包括基于最小二乘的材料分解（LS-MD），全变差正则化材料分解（TV-MD），和一个U-网为基础的方法。与U-net、TV-MD和LS-MD相比，Incept-net提高了基础材料预测质量密度的准确度：在所有含碘插入物（2.0 - 24.0 mgI/cc）中，Incept-net、U-net、TV-MD和LS-MD的碘平均绝对误差（MAE）分别为0.66、1.0、1.33和1.57 mgI/cc。以LS-MD为基准，Incept-net和U-net实现了相当的降噪效果（均在95%左右），均高于TV-MD（85%）。所提出的IGC正则化器有效地帮助Incept-net和U-net减少图像伪影。Incept-net在猪图像中紧密地保持了总质量密度（即质量守恒约束），这验证了其输出在解剖背景中的定量准确性。总的来说，Incept-net的性能比两种传统方法更少依赖于辐射剂量水平;在参数减少约40%的情况下，Incept-net比比较器U-net实现了相对改善的性能，这表明Incept-net的性能增益不是通过简单地增加网络学习能力来实现的。Incept-net表现出比传统方法更好的上级定性图像外观、定量准确性和更低的噪声，并且对剂量变化不敏感。Incep-net推广了看不见的图像结构和不同的材料质量密度，并表现良好。这项研究提供了初步的证据，提出的CNN可以用来提高多能量CT中的材料分解质量。
To develop a convolutional neural network (CNN) that can directly estimate material density distribution from multi-energy CT images without performing conventional material decomposition. The proposed CNN (denoted as Incept-net) followed the general framework of encoder-decoder network, with an assumption that local image information was sufficient for modeling the non-linear physical process of multi-energy CT. Incept-net was implemented with a customized loss function, including an in-house-designed image-gradient-correlation (IGC) regularizer to improve edge preservation. The network consisted of two types of customized multi-branch modules exploiting multi-scale feature representation to improve the robustness over local image noise and artifacts. Inserts with various densities of different materials (hydroxyapatite (HA), iodine, a blood-iodine mixture, and fat) were scanned using a research photon-counting-detector (PCD) CT with two energy thresholds and multiple radiation dose levels. The network was trained using phantom image patches only, and tested with different-configurations of full field-of-view phantom and in vivo porcine images. Further, the nominal mass densities of insert materials were used as the labels in CNN training, which potentially provided an implicit mass conservation constraint. The Incept-net performance was evaluated in terms of image noise, detail preservation, and quantitative accuracy. Its performance was also compared to common material decomposition algorithms including least-square-based material decomposition (LS-MD), total-variation regularized material decomposition (TV-MD), and a U-net based method. Incept-net improved accuracy of the predicted mass density of basis materials compared with the U-net, TV-MD, and LS-MD: the mean absolute error (MAE) of iodine was 0.66, 1.0, 1.33, and 1.57 mgI/cc for Incept-net, U-net, TV-MD and LS-MD, respectively, across all iodine-present inserts (2.0 to 24.0 mgI/cc). With the LS-MD as the baseline, Incept-net and U-net achieved comparable noise reduction (both around 95%), both higher than TV-MD (85%). The proposed IGC regularizer effectively helped both Incept-net and U-net to reduce image artifact. Incept-net closely conserved the total mass densities (i.e. mass conservation constraint) in porcine images, which heuristically validated the quantitative accuracy of its outputs in anatomical background. In general, Incept-net performance was less dependent on radiation dose levels than the two conventional methods; with approximately 40% less parameters, the Incept-net achieved relatively improved performance than the comparator U-net, indicating that performance gain by Incept-net was not achieved by simply increasing network learning capacity. Incept-net demonstrated superior qualitative image appearance, quantitative accuracy, and lower noise than the conventional methods and less sensitive to dose change. Incept-net generalized and performed well with unseen image structures and different material mass densities. This study provided preliminary evidence that the proposed CNN may be used to improve the material decomposition quality in multi-energy CT.
登录
查看更多内容
全身研究光子计数检测器CT的光谱性能：派生图像集中的定量精度。
影响因子：
3.5
作者：
Leng S;Zhou W;Yu Z;Halaweish A;Krauss B;Schmidt B;Yu L;Kappler S;McCollough C
通讯作者：
McCollough C
影响因子：
5
作者：
Chandarana, Hersh;Megibow, Alec J.;Macari, Michael
通讯作者：
Macari, Michael
影响因子：
3.8
作者：
Lu, Yanye;Kowarschik, Markus;Maier, Andreas
通讯作者：
Maier, Andreas
影响因子：
3.8
作者：
KALENDER, WA;PERMAN, WH;KLOTZ, E
通讯作者：
KLOTZ, E
影响因子：
10.6
作者：
Jin, Kyong Hwan;McCann, Michael T.;Unser, Michael
通讯作者：
Unser, Michael