CT-based attenuation correction in the calculation of semi-quantitative indices of [18F]FDG uptake in PET.

The introduction of combined PET/CT systems has a number of advantages, including the utilisation of CT images for PET attenuation correction (AC). The potential advantage compared with existing methodology is less noisy transmission maps within shorter times of acquisition. The objective of our investigation was to assess the accuracy of CT attenuation correction (CTAC) and to study resulting bias and signal to noise ratio (SNR) in image-derived semi-quantitative uptake indices. A combined PET/CT system (GE Discovery LS) was used. Different size phantoms containing variable density components were used to assess the inherent accuracy of a bilinear transformation in the conversion of CT images to 511 keV attenuation maps. This was followed by a phantom study simulating tumour imaging conditions, with a tumour to background ratio of 5:1. An additional variable was the inclusion of contrast agent at different concentration levels. A CT scan was carried out followed by 5 min emission with 1-h and 3-min transmission frames. Clinical data were acquired in 50 patients, who had a CT scan under normal breathing conditions (CTAC(nb)) or under breath-hold with inspiration (CTAC(insp)) or expiration (CTAC(exp)), followed by a PET scan of 5 and 3 min per bed position for the emission and transmission scans respectively. Phantom and patient studies were reconstructed using segmented AC (SAC) and CTAC. In addition, measured AC (MAC) was performed for the phantom study using the 1-h transmission frame. Comparing the attenuation coefficients obtained using the CT- and the rod source-based attenuation maps, differences of 3% and <6% were recorded before and after segmentation of the measured transmission maps. Differences of up to 6% and 8% were found in the average count density (SUV(avg)) between the phantom images reconstructed with MAC and those reconstructed with CTAC and SAC respectively. In the case of CTAC, the difference increased up to 27% with the presence of contrast agent. The presence of metallic implants led to underestimation in the surrounding SUV(avg) and increasing non-uniformity in the proximity of the implant. The patient study revealed no statistically significant differences in the SUV(avg) between either CTAC(nb) or CTAC(exp) and SAC-reconstructed images. The larger differences were recorded in the lung. Both the phantom and the patient studies revealed an average increase of approximately 25% in the SNR for the CTAC-reconstructed emission images compared with the SAC-reconstructed images. In conclusion, CTAC(nb) or CTAC(exp) is a viable alternative to SAC for whole-body studies. With CTAC, careful consideration should be given to interpretation of images and use of SUVs in the presence of oral contrast and in the proximity of metallic implants.

PET attenuation coefficients from CT images: experimental evaluation of the transformation of CT into PET 511-keV attenuation coefficients.

The CT data acquired in combined PET/CT studies provide a fast and essentially noiseless source for the correction of photon attenuation in PET emission data. To this end, the CT values relating to attenuation of photons in the range of 40-140 keV must be transformed into linear attenuation coefficients at the PET energy of 511 keV. As attenuation depends on photon energy and the absorbing material, an accurate theoretical relation cannot be devised. The transformation implemented in the Discovery LS PET/CT scanner (GE Medical Systems, Milwaukee, Wis.) uses a bilinear function based on the attenuation of water and cortical bone at the CT and PET energies. The purpose of this study was to compare this transformation with experimental CT values and corresponding PET attenuation coefficients. In 14 patients, quantitative PET attenuation maps were calculated from germanium-68 transmission scans, and resolution-matched CT images were generated. A total of 114 volumes of interest were defined and the average PET attenuation coefficients and CT values measured. From the CT values the predicted PET attenuation coefficients were calculated using the bilinear transformation. When the transformation was based on the narrow-beam attenuation coefficient of water at 511 keV (0.096 cm(-1)), the predicted attenuation coefficients were higher in soft tissue than the measured values. This bias was reduced by replacing 0.096 cm(-1) in the transformation by the linear attenuation coefficient of 0.093 cm(-1) obtained from germanium-68 transmission scans. An analysis of the corrected emission activities shows that the resulting transformation is essentially equivalent to the transmission-based attenuation correction for human tissue. For non-human material, however, it may assign inaccurate attenuation coefficients which will also affect the correction in neighbouring tissue.

Influence of arm positioning on tomographic thallium-201 myocardial perfusion imaging and the effect of attenuation correction.

Lateral attenuation in single-photon emission tomography (SPET) myocardial perfusion imaging (MPI) has been attributed to the left arm if it is held by the patient's side during data acquisition. As a result MPI data are conventionally acquired with the arms held above the head. The aims of this study were to determine the effect of imaging arms down on reconstructed tomographic images depicting regional myocardial thallium-201 distribution and to assess whether attenuation-corrected (AC) myocardial perfusion images acquired arms down could replace uncorrected (NC) images acquired arms up for routine clinical service. Twenty-eight patients referred for routine MPI underwent sequential 180 degrees emission/transmission imaging for attenuation correction using an L-shaped dual-headed gamma camera (GE Optima) fitted with two gadolinium-153 scanning line sources. Delay data were acquired twice: once supine with the arms up and then supine with the arms down. Detector radius of rotation (ROR) for arms up and arms-down studies was recorded. For each data set, count density was measured in 17 segments of a polar plot and segmental uptake expressed relative to study maximum. Oblique images were assessed qualitatively by two observers blinded to study type for tracer distribution and overall quality. Transmission maps were assessed for truncation. Mean detector ROR was 190 mm for arms-up studies and 232 mm for arms-down studies (P<0.05). Population mean segmental relative uptake values for NC arms-up studies were higher than for NC arms-down studies, with the greatest difference seen anterolaterally. Nevertheless, the majority (24/28) of oblique NC arms-up and NC arms-down images appeared similar and only four (14%) NC arms-down studies showed additional areas of reduced count density (one anterior and three lateral). Corresponding AC arms-down studies showed that count density within the anterior defect improved to normal but the lateral reductions persisted, and in two of these three studies the arms-down transmission map was distorted. Population mean segmental relative uptake values for NC arms-down studies were lower than for AC arms-down studies apart from three anterolateral segments where NC arms-down values were higher. Of 28 AC arms-down studies, 11 (39%) were of reduced quality compared with NC arms-up studies because of poorer spatial resolution and because AC enhances liver activity compared with NC. It is concluded that arm positioning influences reconstructed tomographic images depicting regional 201T1 distribution, particularly anterolaterally. There is lateral undercorrection in approximately 10% of AC arms-down studies, possibly because of attenuation map truncation. Image quality is reduced in about one-third of AC armsdown studies compared with NC arms-up studies. These data suggest that this attenuation correction method is not sufficiently robust to allow routine acquisition of MPI data with the arms down.

Transmission scanning for attenuation correction of myocardial 201Tl images in obese patients.

For attenuation correction (AC) of 201Tl myocardial perfusion images, an accurate attenuation map is required. This study assessed whether prolonged transmission scanning is required in obese compared to normal-sized patients. Twenty-nine obese patients (mean body mass index 33 kg m-2) underwent sequential emission/transmission imaging for AC using an L-shaped, dual-headed gamma camera fitted with two 153Gd scanning line sources. Transmission data were acquired for 5 s per view (scan time for normal-sized patients) and for 10 s per view and used to reconstruct individual attenuation maps. Emission data were reconstructed using each attenuation map in turn to produce attenuation-corrected images (AC5 and AC10). Tracer distribution in the AC5 and AC10 images was compared by two observers blinded to study type. For each data set, count density was measured in 17 segments of a polar plot and segmental uptake expressed relative to study maximum. Although myocardial count density was low on the 5 s per view transmission images (0.5-13.0 and 3.0-14.0 counts per pixel in the anteroposterior and lateral projections respectively), no significant differences in tracer distribution were seen between the AC5 and AC10 images and these were reported identically. In addition, the mean segmental relative uptake values were similar (P > 0.05) for corresponding segments of the AC5 and AC10 images. We conclude that prolonged transmission scanning is not required in obese compared to normal-sized patients. The transmission scanning protocol used in normal-sized patients is applicable across a wide patient weight range.

Effect of attenuation correction on myocardial thallium-201 distribution in patients with a low likelihood of coronary artery disease.

Regional variation of tracer distribution is seen in uncorrected thallium-201 images of normal hearts. This study evaluates the effect of attenuation correction on myocardial 201T1 distribution in patients with low risk of coronary artery disease. An L-shaped dual-detector single-photon emission tomographic system equipped with a pair gadolinium-153 scanning line sources was used for sequential emission/transmission imaging in 36 patients (14 men and 22 women) with less than 5% risk for coronary artery disease. Uncorrected emission images were reconstructed using filtered back-projection (FBP) whereas the attenuation corrected (AC) images were iteratively reconstructed using the attenuation map computed from the transmission data. Both sets of images were reorientated into short axis, vertical long axis and horizontal long axis images. For quantification data were reconstructed into polar plots and count density estimated in 17 myocardial segments. The population % standard deviation for each segment of AC data was significantly smaller than that for FBP data, indicating improved homogeneity of tracer distribution. In men the anterior-basal inferior activity ratio improved from 1.20 for FBP to 0.96 for AC (stress) and from 1.23 for FBP to 0.98 for AC (delay) (P < 0.0001). In women the anterior-basal inferior activity ratio changed from 1.08 for FBP to 0.94 for AC (stress) and from 1.08 for FBP to 0.93 for AC (delay) (P < 0.001). These ratios reflect appropriate compensation for basal attenuation but a lack of scatter correction. The lateral-septal activity ratio in men changed from 1.05 for FBP to 0.99 for AC (stress) and from 1.02 for FBP to 0.96 for AC (delay), while in women it changed from 1.05 for FBP to 0.98 for AC (stress) and from 1.04 for FBP to 0.98 for AC (delay) (P < 0.005 in all cases). The apex of AC images showed a decrease in activity consistent with wall thining at this site. It is concluded that the use of attenuation correction yields improved homogeneity of myocardial tracer distribution in patients with low risk of coronary artery disease. The diagnostic benefits of attenuation correction are yet to be fully assessed.