Huxohl Tamino, Patel Gopesh, Zabel Reinhard, Burchert Wolfgang
Institute of Radiology, Nuclear Medicine and Molecular Imaging, Heart and Diabetes Center North Rhine-Westphalia, University Hospital of the Ruhr University Bochum, Bad Oeynhausen, Germany.
Institute of Nuclear Medicine, Hospital Lippe, Lippe, Germany.
EJNMMI Phys. 2023 Aug 28;10(1):49. doi: 10.1186/s40658-023-00568-1.
The use of CT images for attenuation correction of myocardial perfusion imaging with single photon emission computer tomography (SPECT) increases diagnostic confidence. However, acquiring a CT image registered to a SPECT image is often not possible because most scanners are SPECT-only. It is possible to approximate attenuation maps using deep learning, but this has not yet been shown to work for a SPECT scanner with an IQSPECT collimator. This study investigates whether it is possible to approximate attenuation maps from non-attenuation-corrected (nAC) reconstructions acquired with a SPECT scanner equipped with an IQSPECT collimator.
Attenuation maps and reconstructions were acquired retrospectively for 150 studies. A U-Net was trained to predict attenuation maps from nAC reconstructions using the conditional generative adversarial network framework. Predicted attenuation maps are compared to real attenuation maps using the normalized mean absolute error (NMAE). Attenuation-corrected reconstructions were computed, and the resulting polar maps were compared by pixel and by average perfusion per segment using the absolute percent error (APE). The training and evaluation code is available at https://gitlab.ub.uni-bielefeld.de/thuxohl/mu-map .
Predicted attenuation maps are similar to real attenuation maps, achieving an NMAE of 0.020±0.007. The same is true for polar maps generated from reconstructions with predicted attenuation maps compared to CT-based attenuation maps. Their pixel-wise absolute distance is 3.095±3.199, and the segment-wise APE is 1.155±0.769.
It is feasible to approximate attenuation maps from nAC reconstructions acquired by a scanner with an IQSPECT collimator using deep learning.
使用CT图像对单光子发射计算机断层扫描（SPECT）心肌灌注成像进行衰减校正可提高诊断可信度。然而，获取与SPECT图像配准的CT图像通常是不可能的，因为大多数扫描仪仅具备SPECT功能。利用深度学习可以近似衰减图，但尚未证明其对配备IQSPECT准直器的SPECT扫描仪有效。本研究调查了是否可以从配备IQSPECT准直器的SPECT扫描仪获取的非衰减校正（nAC）重建图像中近似衰减图。
对150项研究进行回顾性采集衰减图和重建图像。使用条件生成对抗网络框架训练一个U-Net，以从nAC重建图像中预测衰减图。使用归一化平均绝对误差（NMAE）将预测的衰减图与真实衰减图进行比较。计算衰减校正后的重建图像，并使用绝对百分比误差（APE）按像素和按每段平均灌注对所得的极坐标图进行比较。训练和评估代码可在https://gitlab.ub.uni-bielefeld.de/thuxohl/mu-map获取。
预测的衰减图与真实衰减图相似，NMAE为0.020±0.007。与基于CT的衰减图相比，使用预测衰减图重建生成的极坐标图也是如此。它们的逐像素绝对距离为3.095±3.199，逐段APE为1.155±0.769。
使用深度学习从配备IQSPECT准直器的扫描仪获取的nAC重建图像中近似衰减图是可行的。
