García-Pérez Pablo, España Samuel
Departamento de Estructura de la Materia, Física Térmica y Electrónica, Facultad de Ciencias Físicas, Universidad Complutense de Madrid, IdISSC, Ciudad Universitaria, 28040, Madrid, Spain.
Centro Nacional de Investigaciones Cardiovasculares (CNIC), Madrid, Spain.
EJNMMI Phys. 2020 Jan 13;7(1):3. doi: 10.1186/s40658-020-0272-9.
Simultaneous reconstruction of emission and attenuation images in time-of-flight (TOF) positron emission tomography (PET) does not provide a unique solution. In this study, we propose to solve this limitation by including additional information given by a reference object with known attenuation placed outside the patient. Different configurations of the reference object were studied including geometry, material composition, and activity, and an optimal configuration was defined. In addition, this configuration was tested for different timing resolutions and noise levels.
The proposed strategy was tested in 2D simulations obtained by forward projection of available PET/CT data and noise was included using Monte Carlo techniques. Obtained results suggest that the optimal configuration corresponds to a water cylinder inserted in the patient table and filled with activity. In that case, mean differences between reconstructed and true images were below 10%. However, better results can be obtained by increasing the activity of the reference object.
This study shows promising results that might allow to obtain an accurate attenuation map from pure TOF-PET data without prior knowledge obtained from CT, MRI, or transmission scans.
在飞行时间(TOF)正电子发射断层扫描(PET)中同时重建发射图像和衰减图像并不能提供唯一解。在本研究中,我们建议通过纳入放置在患者体外具有已知衰减的参考物体所提供的额外信息来解决这一局限性。研究了参考物体的不同配置,包括几何形状、材料组成和活度,并定义了最佳配置。此外,还针对不同的时间分辨率和噪声水平对该配置进行了测试。
所提出的策略在通过对可用PET/CT数据进行正向投影获得的二维模拟中进行了测试,并使用蒙特卡罗技术纳入了噪声。所得结果表明,最佳配置对应于插入患者检查床并充满活度的水缸。在这种情况下,重建图像与真实图像之间的平均差异低于10%。然而,通过增加参考物体的活度可以获得更好的结果。
本研究显示出有前景的结果,这可能允许从纯TOF-PET数据中获得准确的衰减图,而无需从CT、MRI或透射扫描中获取先验知识。