Direct and indirect strategies of deep-learning-based attenuation correction for general purpose and dedicated cardiac SPECT.
Direct and indirect strategies of deep-learning-based attenuation correction for general purpose and dedicated cardiac SPECT.
复制标题
用于通用和专用心脏单光子发射计算机断层成像（SPECT）的基于深度学习的衰减校正的直接和间接策略
DOI：
10.1007/s00259-022-05718-8
复制
发表时间：
2022-07
影响因子：
9.1
通讯作者：
Liu, Chi
中科院分区：
文献类型：
--
作者：
Chen, Xiongchao;Zhou, Bo;Xie, Huidong;Shi, Luyao;Liu, Hui;Holler, Wolfgang;Lin, MingDe;Liu, Yi-Hwa;Miller, Edward J.;Sinusas, Albert J.;Liu, Chi
SPECT的基于深度学习的衰减校正（AC）包括间接和直接方法。间接方法从发射图像生成衰减图（μ图），而直接方法直接从没有μ图的非衰减校正（NAC）图像预测AC图像。对于具有CZT探测器的专用心脏SPECT扫描仪，由于有限的视场（FOV），间接方法具有挑战性。在这项工作中，我们的目标是：1）首先开发新的间接方法来提高专用SPECT的AC性能; 2）比较通用和专用SPECT的直接和间接方法之间的AC性能。对于专用SPECT，我们开发了策略来预测使用小矩阵重建的NAC图像的截断μ图，或使用GE Discovery NM/CT 570 c SPECT/CT扫描的270项匿名临床研究使用大矩阵重建的NAC图像的完整μ图。对于通用SPECT，我们使用在GE NM/CT 850 c SPECT/CT上扫描的400项匿名临床研究实施了直接和间接方法。分别输入NAC图像的峰值和散射窗，预测μ-map或AC图像。对于专用SPECT，使用我们提出的具有完整μ图的策略的平均归一化均方误差（NMSE）为1.20 ± 0.72%，而使用先前的直接方法的平均归一化均方误差（NMSE）为2.21 ± 1.17%。使用我们的方法的极坐标图绝对百分比误差（APE）为3.24 ± 2.79%（R2 = 0.9499），而使用直接方法的APE为4.77 ± 3.96%（R2 = 0.9213）。对于通用SPECT，使用直接方法预测AC图像的平均NMSE为2.57 ± 1.06%，而使用间接方法为1.37 ± 1.16%。我们开发了生成具有小FOV的专用心脏SPECT的μ图的策略。对于通用和专用SPECT，间接方法显示AC的性能上级直接方法。
Deep-learning-based attenuation correction (AC) for SPECT include both indirect and direct approaches. Indirect approaches generate attenuation maps (μ-maps) from emission images, while direct approaches predict AC images directly from non-attenuation-corrected (NAC) images without μ-maps. For dedicated cardiac SPECT scanners with CZT detectors, indirect approaches are challenging due to the limited field-of-view (FOV). In this work, we aim to 1) first develop novel indirect approaches to improve the AC performance for dedicated SPECT; and 2) compare the AC performance between direct and indirect approaches for both general purpose and dedicated SPECT. For dedicated SPECT, we developed strategies to predict truncated μ-maps from NAC images reconstructed with a small matrix, or full μ-maps from NAC images reconstructed with a large matrix using 270 anonymized clinical studies scanned on a GE Discovery NM/CT 570c SPECT/CT. For general purpose SPECT, we implemented direct and indirect approaches using 400 anonymized clinical studies scanned on a GE NM/CT 850c SPECT/CT. NAC image in both photopeak and scatter windows were input to predict μ-maps or AC images. For dedicated SPECT, the averaged normalized mean square error (NMSE) using our proposed strategies with full μ-maps was 1.20 ± 0.72% as compared to 2.21 ± 1.17% using the previous direct approaches. The polar map absolute percent error (APE) using our approaches was 3.24 ± 2.79% (R2 = 0.9499) as compared to 4.77 ± 3.96% (R2 = 0.9213) using direct approaches. For general purpose SPECT, the averaged NMSE of the predicted AC images using the direct approaches was 2.57 ± 1.06% as compared to 1.37 ± 1.16% using the indirect approaches. We developed strategies of generating μ-maps for dedicated cardiac SPECT with small FOV. For both general purpose and dedicated SPECT, indirect approaches showed superior performance of AC than direct approaches.
登录
查看更多内容
影响因子：
4
作者：
Liu F;Jang H;Kijowski R;Zhao G;Bradshaw T;McMillan AB
通讯作者：
McMillan AB
影响因子：
9.3
作者：
Hwang, Donghwi;Kang, Seung Kwan;Lee, Jae Sung
通讯作者：
Lee, Jae Sung
DOI：
10.1007/s00259-009-1143-8
发表时间：
2009-10-01
影响因子：
9.1
作者：
Nesterov, Sergey V.;Han, Chunlei;Knuuti, Juhani
通讯作者：
Knuuti, Juhani
DOI：
10.1007/s12652-020-02827-w
发表时间：
2021-01-03
影响因子：
--
作者：
Liu, Hongqing
通讯作者：
Liu, Hongqing
影响因子：
2.4
作者：
Liu, YH;Sinusas, AJ;Wackers, FJT
通讯作者：
Wackers, FJT