[en] The purpose of this study was to design a method for optimal filter selection during the reconstruction of clinical SPECT images. Hamming, Bartlett, Parzen and Butterworth filters were evaluated at different cutoff frequencies when applied to reconstruction of the Jaszczak phantom and liver SPECTs. The phantom filled with 6 mCi of Tc-99m was imaged following 4 different protocols which varied in matrix sizes (128 x 128 or 64 x 64) and in number of steps (128 or 64). Total imaging time in the 4 protocols was 24 minutes. A total of 160 reconstructions were analyzed. Liver SPECTs from 2 patients with small metastatic lesions from colon Ca were similarly studied. An ECT Performance Index (ECT PI) was defined as the product of the contrast efficiency function (ECT C) and uniformity (ECT U). ECT C as a function of the radius was measured following Rollo's approach. ECT U was measured as the ratio between min. and max. counts per pixel in a known uniform region. ECT PI was computed on a slice through the void spheres region of the phantom. In liver SPECTs the ECT U was measured over the spleen. The most favorable ECT PI (0.35, radius 7.9 mm) was obtained with images in 128 x 128 matrices, 128 steps, processed with a Butterworth cutoff frequency of 0.19, filter order 4. When images were acquired in 64 x 64 matrices using 64 steps the ECT PI was lower and influenced to a lesser degree by both choice of filter and cutoff frequency. Results in the two liver SPECT examinations were parallel to those found in the phantom studies confirming the clinical usefulness of the ECT PI in the evaluation of filters for reconstruction of SPECT images

[en] Lung endothelial N-isopropyl-p-iodoamphetamine (IMP) binding sites were assessed applying principles of competitive binding assay adapted for in vivo measurements obtained by digital imaging. Iodine-¹²³ (123I) IMP, the test cellular tracer, and technetium-99m (/sup 99m/Tc) dextran, the reference vascular tracer were imaged during their first pass through the superior vena cava, right heart, lungs, and left heart in West African dwarf goats. The lung fractional extraction of IMP diminished progressively from 0.96 to 0.20 as the amount of IMP in the test tracer boluses was gradually increased from 0.6 to 150 mg. This demonstrated that lung extraction of IMP is by way of a saturable binding system, presumably receptors. The amount of IMP bound at saturation (R), was found to be 30 mg. Assuming that a single molecule of IMP bound a single receptor, the total number of free receptors was computed as the Avogadro's number times R, divided by the IMP molecular weight, and found to be 6.04 X 10(19). Using a computer model, it was determined that the 20 mg per bolus isotherm was the most sensitive for measuring the number of total free receptors (binding sites). This is the first time, to our knowledge, that noninvasive in vivo assessment of receptors in lung has been accomplished

[en] The purpose of this research was to: a) evaluate variations in sensitivity and uniformity of SPECT detectors during 360⁰ rotation, b) explore the causes of the variations and c) to discuss a correction procedure. A flood source consisting of a lucite disc 47.7 cm in diameter containing 3 microcuries of Co-57 was constructed. This source can be firmly attached to the uncollimated detector guaranteeing no detector-source geometry change during rotation. Four different SPECT cameras were tested. Measurements were obtained at 45⁰ intervals throughout a 360⁰ rotation, and repeated in 3 different orientations with respect to the earth's magnetic field. In one camera the effects of the direction and strength of induced magnetic fields were studied. All detectors showed cyclic rotational variations in sensitivity; in three cameras 6%, in one 3%. The amplitude and phase of the sensitivity variation curves and the variations in uniformity were unique for each camera. The detector's degree of variation did not change with regard to the earth's magnetic field orientation, but did change with regard to direction and strength of B fields. By measuring the gain of each PM tube at different detector positions, with and without B field influence, it was found that the gains shift were the cause of rotational changes in sensitivity and uniformity. By recording the variations of the detector during rotation at the clinical setting, a correction factor (either using software or hardware) can be applied to each angular image to improve SPECT resolution

No abstract available

[en] Recently, the lung has received increasing attention as a metabolic organ. In this role, the lung modulates the composition of the arterial blood by several mechanisms: removing active substances from the plasma, releasing substances into the plasma, temporarily holding substances from circulation, and activating or inactivating substances that pass through the lungs. In this report, the procedures proposed by different investigators for in vivo noninvasive assessment of the lung metabolic functions are reviewed. Most procedures are based on an estimation of the clearance of plasma amines by the lung endothelial cells. This clearance is assessed by measuring the lung uptake or the extraction fraction of an intravenously (IV) injected radiolabeled amine. Our own procedure, which assesses the number of free pulmonary endothelial amine receptors, is discussed in detail. In our procedure, the number of receptors was computed using the number of injected molecules of amine and determining the lung extraction fraction of the amine during its first pass through the lungs. In goats, using N-isopropyl-p-iodoamphetamine labeled with ¹²³I as the radiopharmaceutical, the total number of endothelial lung amine receptors was found to be 1.589 X 10(20). The methods for studying the lung metabolic functions, which are discussed in this report can be applied in humans to evaluate either physiological or pathological conditions

[en] Lung endothelial N-isopropyl-p-iodoamphetamine (IMP) receptors were assessed applying principles of competitive binding assay adapted for in vivo measurements obtained by digital imaging. Data were acquired following a previously published method. I-123=IMP, the test cellular tracer, and Tc-99m-Dexiran, the reference vascular tracer were imaged during their first pass through the SVC, right heart, lungs and left heart in West African dwarf goats. The lung fractional extraction of IMP diminished progressively from 0.96 to 0.20 as the amount of IMP in the test tracer boli was gradually increased from 0.6 to 150 mg. This demonstrated that lung extraction of IMP is via a saturable carrier mediated transport mechanism, receptors. The dissociation constant of IMP-lung binding sites reaction was calculated by Scatchard plot and found to be 11.7 mg. The amount of IMP bound to the receptors at saturation (R), was found to be 30 mg. Assuming that a single molecule of IMP bound a single receptor, the total number of free receptors was computed as the Avogadro's number times R, divided by the IMP molecular weight, and found to be 6.06 x 10E+19. Using a computer model, it was determined that the 12 mg. per bolus isotherm was the most sensitive for measuring the number of total free receptors. It is the first time that noninvasive in vivo assessment of receptors has been accomplished. Basically, the method used can be applied in humans and also to assess receptors in organs other than the lungs