Christina Xing / Physics / Faculty Mentor: Mingwu Jin
Cardiac and respiratory dual-gated single-photon emission computed tomography (SPECT) is a promising technique for minimizing the motion artifacts in myocardial perfusion imaging using SPECT. However, attenuation correction (AC) for dual-gated SPECT using an attenuation map averaged over the respiratory cycle may lead to mis-matched attenuation correction artifacts. In this study, we propose a respiratory motion-matched attenuation correction (RMM-AC) to further improve dual-gated SPECT. For each respiratory gate, RMM-AC uses an attenuation map that matches the SPECT image in 4D reconstruction. Filtered backprojection reconstruction without AC (FBP-nAC) and 4D reconstruction with the attenuation map averaged over all respiratory gates (AVE-AC) are used for comparison. We used the NCAT phantom and SIMIND Monte Carlo simulation to simulate dual-gated cardiac SPECT using 99mTc-sestamibi. An ischemic lesion with 20% uptake reduction at the anterior wall of the left ventricle was used for a channelized Hotelling observer (CHO) study. Three methods were quantitatively evaluated using the root mean squared error (RMSE) and the area under the receiver operating characteristic (ROC) curve (Az) of CHO. RMM-AC outperforms FBP-nAC and AVE-AC in both metrics, which reflects improvements on both reconstruction accuracy and lesion detectability. This study paves the way for the clinical adaptation of dual-gated cardiac SPECT.
Leave a Reply