心肌灌注SPECT仍是临床上診断心脏病的主流方法.呼吸运动是一个主要影响图像品貭的因素.心脏在头脚方向因呼吸的位移可达兩公分,导致明显的图像模糊,心肌活度减少或是量化誤差.此外, SPECT的采集时间長, 是很多个呼吸周期平均的結果, 而CT的采集时间僅為一个呼吸周期中的一小段,用此CT來做SPECT的衰减校正会导致假影與图像錯位..呼吸門控在SPECT中能减轻呼吸所引起的模糊.但是,因著每个呼吸态所偵測的光子变少,图像嗓声相对增加.有学者提出只使用时间較長的吐气末态的数据來做权衡.我們团隊已提出使用內差平均CT作PET/CT的衰减校正.但是,對於心脏SPECT/CT的最佳化CT衰减校正方法目前尚沒有定论..在本計划中, 我們的目标是: 1) 优化呼吸門控的方法; 2)決定呼吸門控SPECT的最合適CT衰减校正图; 3)评估图像配准能否進一步利用整个呼吸态的数据以降低嗓声和提升图像品貭.

Myocardial perfusion single photon emission computed tomography (SPECT) continues to be one of the most established methods for diagnosis of cardiac disease. Respiratory motion is one of the main image degradation factors. It was reported that the amplitude of the myocardial respiration motion in the cranial-caudal direction can reach 2 cm. It can cause substantial image blurring, reduction in myocardial activity in anterior and inferior walls leading to false positive cases and bias in image quantification. Besides, SPECT acquisition spans over extended periods with free breathing while conventional helical CT captures only a partial respiratory cycle, using this CT for attenuation correction results in substantial artifacts..Respiratory gating is proposed to reduce the respiratory motion blurring. Usually a respiratory tracking device is needed to obtain the respiratory signal from the patient. The signal then needs to be synchronized with the list mode data for the later rebinning. However, the noise increases as the photon counts in each respiratory phase are reduced. Researchers proposed to use the end-expiratory phase only for image reconstruction as the patients tend to spend more time in this phase during the whole respiratory cycle. .In PET/CT, researchers proposed using cine average CT or 4D CT for static or gated PET reconstruction. However, radiation dose is the main concern. Our group proposed the interpolated average CT for attenuation correction in PET/CT and showed its merits in both simulation and patients. A low dose and accurate CT protocol is not yet determined for cardiac SPECT/CT. .In this proposal, we aim to (i) optimize the respiratory gating methods, i.e., time gating or amplitude gating, for cardiac SPECT/CT, and (ii) to determine the corresponding CT protocol for attenuation correction. We will then (iii) evaluate the use of non-rigid registration on gated SPECT to see if we can further reduce the noise by utilizing data from the whole respiratory cycle. Computerized simulation study using digital phantoms, physical phantom and patient studies will be performed to validate our proposed respiratory gating method in cardiac SPECT/CT.