Comparison of deep learning synthesis of synthetic CTs using clinical MRI inputs.
Comparison of deep learning synthesis of synthetic CTs using clinical MRI inputs.
复制标题
DOI：
10.1088/1361-6560/abc5cb
复制
发表时间：
2020-12-23
影响因子：
3.5
通讯作者：
McMillan AB
中科院分区：
文献类型：
--
作者：
Massa HA;Johnson JM;McMillan AB
已经对开发用于从MRI输入合成类CT图像的技术产生了实质性兴趣，其在同时PET/MR和放射治疗规划中具有重要应用。深度学习最近显示出解决这个问题的巨大潜力。本研究的目的是研究四种常见临床MRI序列（T1加权梯度回波[T1]、T2加权脂肪抑制快速自旋回波[T2-FatSat]、造影后T1加权梯度回波[T1-Post]和快速自旋回波T2加权液体衰减反转恢复[CUBE-FLAIR]）作为深部CT合成管道输入的能力。在同一天接受MRI和CT扫描的92例受试者中回顾性获得数据。使用仿射配准将患者的MR和CT扫描彼此配准。深度学习模型是一个卷积神经网络编码器-解码器，具有类似于U-net架构的跳过连接和Inception V3启发的块，而不是顺序卷积块。在使用150个epoch和6个批量进行训练后，使用SSIM，PNSR，MAE和Dice Coefficient对模型进行评估。我们发现，每种图像类型都可以获得可行的结果，并且没有任何一种图像类型在所有分析中均具有上级优势。CUBE-FLAIR、T1、T1-Post和T2的全脑合成CT的MAE（单位：HU）分别为51.236 ± 4.504、45.432 ± 8.517、44.558 ± 7.478和45.721 ± 8.7767，这不仅显示了可行性，而且显示了非常令人信服的临床图像结果。基于深度学习的MRI CT图像合成可以使用广泛的输入，这表明可以从广泛的临床输入类型中创建可行的图像。
There has been substantial interest in developing techniques for synthesis of CT like images from MRI inputs, with important applications in simultaneous PET/MR and radiotherapy planning. Deep learning has recently shown great potential for solving this problem. The goal of this research was to investigate the capability of four common clinical MRI sequences (T1-weighted gradient-echo [T1], T2-weighted fat-suppressed fast spin-echo [T2-FatSat], post-contrast T1-weighted gradient-echo [T1-Post], and fast spin-echo T2-weighted fluid-attenuated inversion recovery [CUBE-FLAIR]) as inputs into a deep CT synthesis pipeline. Data were obtained retrospectively in 92 subjects who had undergone an MRI and CT scan on the same day. The patient’s MR and CT scans were registered to one another using affine registration. The deep learning model was a convolutional neural network encoder-decoder with skip connections similar to the U-net architecture and Inception V3 inspired blocks instead of sequential convolution blocks. After training with 150 epochs and a batch size of 6, the model was evaluated using SSIM, PNSR, MAE, and Dice Coefficient. We found that feasible results were attainable for each image type, and no single image type was superior for all analyses. The MAE (in HU) of the resulting synthesized CT in the whole brain was 51.236 ± 4.504 for CUBE-FLAIR, 45.432 ± 8.517 for T1, 44.558 ± 7.478 for T1-Post, and 45.721 ± 8.7767 for T2, showing not only feasible, but also very compelling results on clinical images. Deep learning-based synthesis of CT images from MRI is possible with a wide range of inputs, suggesting that viable images can be created from a wide range of clinical input types.
登录
查看更多内容
影响因子：
3.3
作者：
Jang H;Liu F;Bradshaw T;McMillan AB
通讯作者：
McMillan AB
影响因子：
5.7
作者：
Jenkinson, M;Bannister, P;Smith, S
通讯作者：
Smith, S
DOI：
10.2967/jnumed.116.174169
发表时间：
2016-12
期刊：
Journal of nuclear medicine : official publication, Society of Nuclear Medicine
影响因子：
--
作者：
Fendler WP;Czernin J;Herrmann K;Beyer T
通讯作者：
Beyer T
DOI：
10.1007/s00259-012-2113-0
发表时间：
2012-07-01
影响因子：
9.1
作者：
Samarin, Andrei;Burger, Cyrill;Kuhn, Felix P.
通讯作者：
Kuhn, Felix P.
影响因子：
19.7
作者：
Liu, Fang;Jang, Hyungseok;McMillan, Alan B.
通讯作者：
McMillan, Alan B.