A deep learning method for producing ventilation images from 4DCT: First comparison with technegas SPECT ventilation
A deep learning method for producing ventilation images from 4DCT: First comparison with technegas SPECT ventilation
复制标题
从 4DCT 生成通气图像的深度学习方法：与 technegas SPECT 通气的首次比较
DOI：
10.1002/mp.14004
复制
发表时间：
2020-01-28
期刊：
影响因子：
3.8
通讯作者：
Dai, Jianrong
中科院分区：
文献类型：
--
作者：
Liu, Zhiqiang;Miao, Junjie;Dai, Jianrong
目的本研究的目的是开发一种用于生成四维计算机断层扫描（4DCT）通气成像的深度学习（DL）方法，并评估基于DL的通气成像与单光子发射计算机断层扫描（SPECT）通气成像（SPECT-VI）的准确性。基于DL的方法的性能进行评估，通过比较与密度变化和雅可比（HU和JAC）的方法。材料与方法50例食管癌和肺癌患者接受胸部放射治疗。对于每名患者，在第一次放射治疗之前获得4DCT扫描与99 mTc-Technegas SPECT/CT配对。首先使用MIMvista对4DCT和SPECT/CT进行严格配准，并使用MATLAB将其转换为数据矩阵，然后转移到基于U-net的DL模型，用于将4DCT特征与SPECT-VI关联起来。研究了两种形式的4DCT数据集[（a）十个相位和（B）呼气峰和吸气峰的两个相位]作为输入。十重交叉验证程序被用来评估DL模型的性能。为了进行比较评价，使用HU和JAC方法计算每个患者基于4DCT（CTVI）的特定通气成像。在每个CTVI和相应的SPECT-VI之间的整个肺上评价体素方面的斯皮尔曼相关性。将SPECT-VI和生成的CTVI分割为高、中、低功能肺（HFL、MFL和LFL）区域。还使用骰子相似系数（DSC）评估了每个CTVI对SPECT-VI的相应HFL、MFL和LFL的空间重叠。采用单因素方差分析模型对不同方法测量的肺功能区DSC进行统计学分析。结果CTVIHU、CTVIJAC和CTVIDL（1）/CTVIDL（2）的体素斯皮尔曼r（s）值分别为（0.22 ± 0.31）、（-0.09 ± 0.18）和（0.73 ± 0.16）/（0.71 ± 0.17）。这些结果表明，DL方法与SPECT-VI的相关性最强。使用DSC作为空间重叠度量，我们发现CTVIHU、CTVIJAC和CTVIDL（1）/CTVIDL（2）方法获得的所有患者的平均DSC值分别为（0.45 +/- 0.08）、（0.33 +/- 0.04）和（0.73 +/- 0.09）/（0.71 +/- 0.09）。结果表明，DL法与SPECT-VI法相似度最高，差异显著（P < 10（-7））。结论建立了CTVI的DL制备方法，并与SPECT-VI进行了对比验证。结果表明，与HU和JAC方法相比，DL方法可以得到精度大大提高的CTVI。所产生的通气图像可以更准确并且对于肺功能避免放射治疗和治疗响应建模是有用的。
Purpose The purpose of this study is to develop a deep learning (DL) method for producing four-dimensional computed tomography (4DCT) ventilation imaging and to evaluate the accuracy of the DL-based ventilation imaging against single-photon emission-computed tomography (SPECT) ventilation imaging (SPECT-VI). The performance of the DL-based method is assessed by comparing with density change- and Jacobian-based (HU and JAC) methods. Materials and methods Fifty patients with esophagus or lung cancer who underwent thoracic radiotherapy were enrolled in this study. For each patient, 4DCT scans paired with 99mTc-Technegas SPECT/CT were acquired before the first radiotherapy treatment. 4DCT and SPECT/CT were first rigidly registered using MIMvista and converted to data matrix using MATLAB, and then transferred to a DL model based on U-net for correlating 4DCT features and SPECT-VI. Two forms of 4DCT dataset [(a) ten phases and (b) two phases of peak-exhalation and peak-inhalation] as input are studied. Tenfold cross-validation procedure was used to evaluate the performance of the DL model. For comparative evaluation, HU and JAC methodologies are used to calculate specific ventilation imaging based on 4DCT (CTVI) for each patient. The voxel-wise Spearman's correlation was evaluated over the whole lung between each of CTVI and corresponding SPECT-VI. The SPECT-VI and produced CTVIs were segmented into high, median, and low functional lung (HFL, MFL, and LFL) regions. The spatial overlap of corresponding HFL, MFL, and LFL for each CTVI against SPECT-VI was also evaluated using the dice similarity coefficient (DSC). The averaged DSC of functional lung regions was calculated and statistically analyzed with a one-factor ANONA model among different methods. Results The voxel-wise Spearman r(s) values were (0.22 +/- 0.31), (-0.09 +/- 0.18), and (0.73 +/- 0.16)/(0.71 +/- 0.17) for the CTVIHU, CTVIJAC, and CTVIDL(1)/CTVIDL(2). These results showed the DL method yielded the strongest correlation with SPECT-VI. Using the DSC as the spatial overlap metric, we found that the CTVIHU, CTVIJAC, and CTVIDL(1)/CTVIDL(2) methods achieved averaged DSC values for all patients to be (0.45 +/- 0.08), (0.33 +/- 0.04), and (0.73 +/- 0.09)/(0.71 +/- 0.09), respectively. The results demonstrated that the DL method yielded the highest similarity with SPECT-VI with the prominently significant difference (P < 10(-7)). Conclusions This study developed a DL method for producing CTVI and performed a validation against SPECT-VI. The results demonstrated that DL method can derive CTVI with greatly improved accuracy in comparison to HU and JAC methods. The produced ventilation images can be more accurate and useful for lung functional avoidance radiotherapy and treatment response modeling.