Two CT images at the same slice location of a patient at two breathing states: (A) inspiration and (B) expiration. The heart region was smaller, and both lung regions were larger in (A) than in (B) in the selected slice location. Attenuation correction of the PET data may be compromised by the CT image in (A) as we normally spend more time in expiration than in inspiration.
Two cardiac CT images with contrast injection at the same slice location at (A) end-systole and (B) end-diastole phases. The patient was holding breath during a cardiac CT scan of , , 0.325:1 pitch, and CT gantry rotation. The vertical line segment marks the same location in each image and indicates that the heart beating did not cause a large displacement at the interface between the heart walls and the left lung.
ACT images can be derived by averaging the cine CT images for AC of the PET emission data. Each cine scan covers of anatomy. We use collimation in our GE Discovery PET/CT scanners with an 8- or 16-slice CT scanner. Each reconstruction generates 8 slices due to the 8-slice collimation. After cine scan at a location, the table travels another to bring the next of anatomy into the cine CT scan. After image averaging, a set of CT images for AC of the PET emission data can then be formed by combining the HCT images above and below the thorax and the ACT images of the thorax.
The distribution of (A) the average and (B) the standard deviation of the breath cycles of the 600 patients scanned with the cine CT over a period of two years. In total, 87% of the patients had average breath cycles of less than from the summation of the 5%, 29%, 33% and 20%, corresponding to the breath cycles of 2–3, 3–4, 4–5, and , respectively; 90% of the patients had standard deviations of less than in the period of breath cycle.
Illustration of the segmentation of myocardium in the PET data corrected with ACT (B) and superimposition of the segmentation on the PET data corrected with HCT (A). The segmentation was done on a slice-by-slice basis, and will be used for the quantitative analysis in Fig. 7.
Fused coronal images of the PET data with HCT and ACT data are shown in (A) and (B), respectively. The two images in each row are from the same patient. The images from top to bottom correspond to patients 1–4. The regions with bright color correspond to the regions of the PET data with uptake and the CT data with heart tissues. The regions with light or dark gray color correspond to the regions of the PET data with uptake and the CT data of lung tissues. The PET data corrected with ACT registered better than the PET data corrected with HCT for all four patients.
Average quantification of myocardium from the superior slices toward the head to the inferior slices near the abdomen. Each slice is . The plots of (A)–(D) correspond to the data of patients 1–4 in sequence. Each plot is normalized to the maximum of average quantification value in myocardium in all slices. The PET data corrected with ACT had a larger average quantification over the PET data corrected with HCT. The maximum changes in percentage from (A) to (D) are 42% (slice 6), 50% (slice 1), 31% (slice 6) and 35% (slice 2). The minimum changes in percentage were 24% (slice 18), 16% (slice 21), 5% (slice 15), and 4% (slice 27). The changes were larger in the superior slices than in the inferior slices.
The reformation of the PET data of patient 1 corrected with the HCT and the ACT data. The images from top to bottom are the short axis (SA), the vertical long axis (VLA), and the horizontal long axis (HLA) images. In each reformation, the PET data corrected with the ACT data are at the upper row and the PET data corrected with the HCT data at the lower row. There was an improvement in uniformity at the anterior and the lateral walls when the PET data were corrected with the ACT data compared to the PET data corrected with the HCT data.
Illustration of the difference in quantification of the PET data in (E) between the PET data in (C), corrected with the HCT data in (A), and the PET data in (D), corrected with the ACT data in (B). A segmented region in (E) is also duplicated in the HCT data in (A) and the ACT data in (B). This example demonstrates the cause of misregistration when a part of the lateral walls of myocardium in the PET data are attenuation corrected with the lower attenuation of the lungs when they should have been corrected with the higher attenuation of the heart tissues.
The 20 segments of the myocardium are shown in (A). The average change in percentage in quantification of the four patients from the PET data corrected with the HCT data to the PET data corrected with the ACT data. The uptake of the anterior and the lateral walls increased; however, the uptake in the septal and the inferior walls decreased.
The ACT and the SSCT images of patient 5 with an average breath cycle of . The SSCT images in (A) were taken with one single CT gantry rotation of , and the two images were apart and thick. The corresponding ACT images in (B), obtained by averaging the cine CT images, were averaged from of data collection over eight gantry rotations. Since the averaging took place on the cine CT images of good temporal resolution, the ACT images were almost free of reconstruction artifacts, which were observed in the SSCT images.
Article metrics loading...
Full text loading...