Comparing filtered backprojection and ordered-subsets expectation maximization for small-lesion detection and localization in 67Ga SPECT.
(73/3327)
Iterative reconstruction of SPECT images has recently become clinically available as an alternative to filtered backprojection (FBP). However, there is conflicting evidence on whether iterative reconstruction, such as with the ordered-subsets expectation maximization (OSEM) algorithm, improves diagnostic performance over FBP. The study objective was to determine if the detection and localization of small lesions in simulated thoracic gallium SPECT images are better with OSEM reconstruction than with FBP, both with and without attenuation correction (AC). METHODS: Images were simulated using an analytic projector acting on the mathematic cardiac torso computer phantom. Perfect scatter rejection was assumed. Lesion detection accuracy was assessed using localization receiver operating characteristic methodology. The images were read by 5 nuclear medicine physicians. For each reconstruction strategy and for each observer, data were collected in 2 viewing sessions of 100 images. Two-way ANOVA and, when indicated, the Scheffe multiple comparisons test were applied to check for significant differences. RESULTS: Little difference in the accuracy of detection or localization was seen between FBP with and without AC. OSEM with AC extended the contrast range for accurate lesion detection and localization over that of the other methods investigated. Without AC, no significant difference between OSEM and FBP reconstruction was detected. CONCLUSION: OSEM with AC may improve the detection and localization of thoracic gallium-labeled lesions over FBP reconstruction. (+info)
Relative impact of scatter, collimator response, attenuation, and finite spatial resolution corrections in cardiac SPECT.
(74/3327)
We determined the relative effect of corrections for scatter, depth-dependent collimator response, attenuation, and finite spatial resolution on various image characteristics in cardiac SPECT. METHODS: Monte Carlo simulations and real acquisition of a 99mTc cardiac phantom were performed under comparable conditions. Simulated and acquired data were reconstructed using several correction schemes that combined different methods for scatter correction (3 methods), depth-dependent collimator response correction (frequency-distance principle), attenuation correction (nonuniform Chang correction or within an iterative reconstruction algorithm), and finite spatial resolution correction (use of recovery coefficients). Five criteia were considered to assess the effect of the processing schemes: bull's-eye map (BEM) uniformity, contrast between the left ventricle (LV) wall and the LV cavity, spatial resolution, signal-to-noise ratio (SNR), and percent errors with respect to the known LV wall and liver activities. RESULTS: Similar results were obtained for the simulated and acquired data. Scatter correction significantly improved contrast and absolute quantitation but did not have noticeable effects on BEM uniformity or on spatial resolution and reduced the SNR. Correction for the depth-dependent collimator response improved spatial resolution from 13.3 to 9.5 mm in the LV region, improved absolute quantitation and contrast, but reduced the SNR. Correcting for attenuation was essential for restoring BEM uniformity (78% and 89% without and with attenuation correction, respectively [ideal value being 100%]) and accurate absolute activity quantitation (errors in estimated LV wall and liver activity decreased from 90% without attenuation correction to approximately20% with attenuation correction only). Although accurate absolute activity quantitation was achieved in the liver using scatter and attenuation corrections only, correction for finite spatial resolution was needed to estimate LV wall activity within 10%. CONCLUSION: The respective effects of corrections for scatter, depth-dependent collimator response, attenuation, and finite spatial resolution on different image features in cardiac SPECT were quantified for a specific acquisition configuration. These results give indications regarding the improvements to be expected when using a specific processing scheme involving some or all corrections. (+info)
Validation of the Yale circumferential quantification method using 201Tl and 99mTc: a phantom study.
(75/3327)
The Yale circumferential quantification (Yale CQ) method for quantification of SPECT images has been validated previously using empirically derived correction factors. In the present studies, the Yale CQ method was further validated using 2 SPECT gamma cameras and 2 radioisotopes. METHODS: SPECT images were acquired from cardiac phantoms with multiple fillable inserts to simulate myocardial perfusion defects of varying extents and severities. Seventy phantom configurations were created. One hundred and forty SPECT images (70 with 99mTc and 70 with 201TI) were acquired using a triple-head SPECT camera. SPECT defects were quantified using the Yale CQ method, with incorporation of 99mTc- and 201TI-derived normal databases and correction factors. RESULTS: Quantified phantom SPECT defect sizes acquired with 99mTc correlated well with actual calculated defect sizes (r = 0.96, y = 0.92x - 0.41). Bland-Altman analysis of agreement revealed strong agreement over a wide range of defect sizes, with a mean error of 1.2% and 2 SDs of 5.0%. Overall 201TI SPECT defect sizes also correlated well with actual defect sizes (r = 0.92), but there was a systematic underestimation (y = 0.72x - 0.76). Bland-Altman analysis showed underestimation over the entire range of defect sizes, with a mean error of 3.4% and 2 SDs of 7.5%. Implementation of a normal 201TI phantom database improved accuracy of quantification (r = 0.95, y = 0.87x - 1.36). The addition of 201TI-specific correction factors further improved accuracy (r= 0.94, y = 0.98x - 1.52). Reproducibility of SPECT defect sizes quantification for 99mTc using 2 gamma cameras was excellent (r = 0.98, y = 0.98x + 0.84). CONCLUSION: The Yale CQ SPECT quantification method, using the empirically derived correction factors, provides accurate and reproducible quantification of phantom defects over a wide range of defect sizes. Accurate quantification of 201TI and 99mTc SPECT defect sizes requires radiotracer-specific normal databases. (+info)
Low-dose high-resolution CT of the petrous bone.
(76/3327)
PURPOSE: To show that CT of the petrous bone can be realized using a low-dose technique. MATERIAL: and methods: A high-contrast phantom was scanned with 1.5 mm slice thickness and 60-510 mAs using the reconstruction algorithms standard, bone and edge. In 50 patients, the petrous bone was examined using the standard protocol at 510 mAS. Additionally, selected slices were made at 120 or 210 mAs. The resolution of relevant structures was compared. Phantom studies were repeated on a second CT-device; images of patients scanned with 80 mAs were analyzed in regard to resolution of osseous details. RESULTS: With the first CT-device structures of the phantom up to 0. 5 mm were depicted using 510 mAs and the edge kernel. With 120 mAs and the bone kernel structures of 0.6 mm could be distinguished. Although the same resolution was achieved with 60 mAs and the edge kernel, patient examinations showed a profound image noise. The results achieved with 120 mAs and the bone algorithm, however, were equal to that of 510 mAs. With the second device the same image quality was realized with only 80 mAs. CONCLUSION: CT-examinations of the petrous bone can be effected without loss of diagnostic information using only 15% of the radiation dose used for a standard brain examination. (+info)
Validation of real-time three-dimensional echocardiography for quantifying left ventricular volumes in the presence of a left ventricular aneurysm: in vitro and in vivo studies.
(77/3327)
OBJECTIVES: To validate the accuracy of real-time three-dimensional echocardiography (RT3DE) for quantifying aneurysmal left ventricular (LV) volumes. BACKGROUND: Conventional two-dimensional echocardiography (2DE) has limitations when applied for quantification of LV volumes in patients with LV aneurysms. METHODS: Seven aneurysmal balloons, 15 sheep (5 with chronic LV aneurysms and 10 without LV aneurysms) during 60 different hemodynamic conditions and 29 patients (13 with chronic LV aneurysms and 16 with normal LV) underwent RT3DE and 2DE. Electromagnetic flow meters and magnetic resonance imaging (MRI) served as reference standards in the animals and in the patients, respectively. Rotated apical six-plane method with multiplanar Simpson's rule and apical biplane Simpson's rule were used to determine LV volumes by RT3DE and 2DE, respectively. RESULTS: Both RT3DE and 2DE correlated well with actual volumes for aneurysmal balloons. However, a significantly smaller mean difference (MD) was found between RT3DE and actual volumes (-7 ml for RT3DE vs. 22 ml for 2DE, p = 0.0002). Excellent correlation and agreement between RT3DE and electromagnetic flow meters for LV stroke volumes for animals with aneurysms were observed, while 2DE showed lesser correlation and agreement (r = 0.97, MD = -1.0 ml vs. r = 0.76, MD = 4.4 ml). In patients with LV aneurysms, better correlation and agreement between RT3DE and MRI for LV volumes were obtained (r = 0.99, MD = -28 ml) than between 2DE and MRI (r = 0.91, MD = -49 ml). CONCLUSIONS: For geometrically asymmetric LVs associated with ventricular aneurysms, RT3DE can accurately quantify LV volumes. (+info)
Sensitive detection of mediastinal lymph node metastasis of lung cancer with 11C-choline PET.
(78/3327)
11C-choline and FDG are PET tracers used to visualize various malignancies. In this study, we compared their capabilities in detecting mediastinal lymph node metastasis originating from non-small cell lung cancer (NSCLC). METHODS: Twenty-nine patients with biopsy-proven NSCLC were studied with PET. 11C-choline PET and FDG PET were performed from 5 and 40 min, respectively, after injection of 370 MBq tracer. PET data were analyzed in terms of the standardized uptake value (SUV). After the PET study, the patients underwent lung resection and mediastinal lymph node dissection. The resected specimens were examined pathologically, and the PET data were analyzed in reference to the pathologic data. RESULTS: With 11C-choline, the SUV in metastasis was similar to the SUV in the primary tumor, where the similarity of the SUV was 100% allowing for a 40% difference. With FDG, small metastases were invisible on the PET image. The SUV of FDG in metastasis was much smaller than that in the primary tumor, and the similarity of the SUV was only 19% allowing for a 40% difference. When pathologic findings were used as standards, the sensitivities of 11C-choline PET and FDG PET in detecting mediastinal lymph node metastasis were 100% and 75%, respectively. CONCLUSION: 11C-choline PET was very effective in detecting lymph node metastases in the mediastinum originating from NSCLC, with a sensitivity of 100%. 11C-choline PET promises to be useful not only before surgery but also after surgery. (+info)
False cerebral activation on BOLD functional MR images: study of low-amplitude motion weakly correlated to stimulus.
(79/3327)
BACKGROUND AND PURPOSE: Movements of the participant during blood oxygen level-dependent (BOLD) functional MR imaging cerebral activation studies are known to produce occasionally regions of false activation, especially when these movements are relatively large (>3 mm) and highly correlated with the stimulus. We investigated whether minimal (<1 mm), weakly correlated movements in a controlled functional MR imaging model could produce false activation artifacts that could potentially mimic regions of true activation in size, location, and statistical significance. METHODS: A life-size brain phantom was constructed by embedding vials of a dilute carboxylic acid solution within a gadolinium-doped gelatin mold. Imaging was performed at 1.5 T using a 2D spiral sequence (3,000/5 [TR/TE]; flip angle, 88 degrees; matrix, 64 x 64; field of view, 24 cm; section thickness, 5 mm). Controlled, in-plane, submillimeter movements of the phantom were generated using a pneumatic system and were made to correlate with a hypothetical "boxcar" stimulus over the range 0.31 < r < 0.96. Regions of false activation were sought using standard statistical methods (SPM96) that excluded phantom edges and accounted for spatial extent (regions tested at P < .05, corrected for multiple comparisons). A similar experiment was performed on a resting volunteer. RESULTS: The pneumatic system provided motion control with average in-plane displacements and rotations of 0.74 mm and 0.47 degrees, respectively, in the 18 data sets analyzed. No areas of false activation in the phantom were identified for poorly correlated motions (r < 0.52). Above this level, false activations occurred with increasing frequency, scaling in size and number with the degree of motion correlation. For motions with r > 0.67, areas of false activation were seen in every experiment. For a statistical threshold of P = .001, the median number of falsely activated regions was 3.5, with a mean size of 71.7 voxels (approximately 5 cc). Areas of possibly false activation of average size 72.5 voxels resulting from passive motion of the resting human participant were observed in two of four experiments. CONCLUSION: Participant movements of 1 mm or less that are only modestly correlated with a blocked stimulus paradigm can produce appreciable false activation artifacts on BOLD functional MR imaging studies, even when strict image realignment methods are used to prevent them. (+info)
Accuracy of 131I tumor quantification in radioimmunotherapy using SPECT imaging with an ultra-high-energy collimator: Monte Carlo study.
(80/3327)
Accuracy of 131I tumor quantification after radioimmunotherapy (RIT) was investigated for SPECT imaging with an ultra-high-energy (UHE) collimator designed for imaging 511-keV photons. METHODS: First, measurements and Monte Carlo simulations were carried out to compare the UHE collimator with a conventionally used, high-energy collimator. On the basis of this comparison, the UHE collimator was selected for this investigation, which was carried out by simulation of spherical tumors in a phantom. Reconstruction was by an expectation-maximization algorithm that included scatter and attenuation correction. Keeping the tumor activity constant, simulations were carried out to assess how volume-of-interest (VOI) counts vary with background activity, radius of rotation (ROR), tumor location, and size. The constant calibration factor for quantification was determined from VOI counts corresponding to a 3.63-cm-radius sphere of known activity. Tight VOIs corresponding to the physical size of the spheres or tumors were used. RESULTS: Use of the UHE collimator resulted in a large reduction in 131I penetration, which is especially significant in RIT where background uptake is high. With the UHE collimator, typical patient images showed an improvement in contrast. Considering the desired geometric events, sensitivity was reduced, but only by a factor of 1.6. Simulation results for a 3.63-cm-radius tumor showed that VOI counts vary with background, location, and ROR by less than 3.2%, 3%, and 5.3%, respectively. The variation with tumor size was more significant and was a function of the background. Good quantification accuracy (<6.5% error) was achieved when tumor size was the same as the sphere size used in the calibration, irrespective of the other parameters. For smaller tumors, activities were underestimated by up to -15% for the 2.88-cm-radius sphere, -23% for the 2.29-cm-radius sphere, and -47% for the 1.68-cm-radius sphere. CONCLUSION: Reasonable accuracy can be achieved for VOI quantification of 131I using SPECT with an UHE collimator and a constant calibration factor. Difference in tumor size relative to the size of the calibration sphere had the biggest effect on accuracy, and recovery coefficients are needed to improve quantification of small tumors. (+info)