An investigation of the magnitude and causes of count loss artifacts in SPECT imaging.

A quantitative evaluation and an investigation of the mechanism of the count losses that can occur in SPECT imaging is described. The most common clinical example of the artifacts which result from this is encountered when sections are taken through the femoral heads in skeletal studies of the pelvis. From some simple phantom studies, it was identified that the count losses were associated with the presence of a high dynamic range in the projection data. Further information was obtained from simulation studies. It was found that the count loss phenomenon is caused by the combined effects of the presence of a high count density area and a structure with a relatively high attenuation level. An appreciation of the presence, magnitude, and mechanism of this effect is important if erroneous clinical findings are to be avoided. This is particularly relevant because of the existence of a variety of techniques that can be used to remove the resulting artifacts.

Detection of a false left ventricular aneurysm by first-pass radionuclide ventriculography.

A false left ventricular aneurysm complicating an inferior myocardial infarction was successfully identified by computerized first-pass radionuclide ventriculography using a multicrystal gamma camera. The aneurysmal chamber exhibited paradoxical systolic expansion, a pattern of contraction confirmed by contrast cineangiography. Because of the propensity of false aneurysms to rupture, early noninvasive firm diagnosis is desirable and may be accomplished by first-pass radionuclide angiography.

Nonistropic point spread function as a result of collimator design and manufacturing defects.

A point spread function (PSF) which was of an unusual form was observed when using a low energy collimator and a 99mTc point source. A lack of radial symmetry and the detection of events over 20 cm away from the center of the source were noted. The major quantitative effect of this was a variation in the resolution of the system with direction. The consequences of this are likely to be of importance especially for single photon emission computed tomographic (SPECT) imaging. The effect was produced by a combination of faults in the design and manufacture of the low energy collimators. It demonstrates the importance of making quantitative measurements of collimator performance in the quality control assessment of gamma camera systems. For SPECT imaging the replacement of lead foil collimators with cast collimators should be considered.

Digital filtering of the bladder in SPECT bone studies of the pelvis.

A data processing technique for the removal of bladder activity from single photon emission computed tomographic bone studies of the pelvis has been developed. The method involves the replacement of count values in the bladder on all projection views by data which are representative of the activity in surrounding structures. Reconstruction is then performed using the amended set of projection views. The method was tested by examining a group of 13 patients referred for investigation of avascular necrosis of the femoral head. Significant improvements in image quality were observed, particularly with respect to the level of artifact production, which increased the number of cases in which a confident and correct diagnosis was made.

Automatic detection of diabetic retinopathy using an artificial neural network: a screening tool.

AIMS: To determine if neural networks can detect diabetic features in fundus images and compare the network against an ophthalmologist screening a set of fundus images. METHODS: 147 diabetic and 32 normal images were captured from a fundus camera, stored on computer, and analysed using a back propagation neural network. The network was trained to recognise features in the retinal image. The effects of digital filtering techniques and different network variables were assessed. 200 diabetic and 101 normal images were then randomised and used to evaluate the network's performance for the detection of diabetic retinopathy against an ophthalmologist. RESULTS: Detection rates for the recognition of vessels, exudates, and haemorrhages were 91.7%, 93.1%, and 73.8% respectively. When compared with the results of the ophthalmologist, the network achieved a sensitivity of 88.4% and a specificity of 83.5% for the detection of diabetic retinopathy. CONCLUSIONS: Detection of vessels, exudates, and haemorrhages was possible, with success rates dependent upon preprocessing and the number of images used in training. When compared with the ophthalmologist, the network achieved good accuracy for the detection of diabetic retinopathy. The system could be used as an aid to the screening of diabetic patients for retinopathy.

Functional imaging of the retina using the multifocal electroretinograph: a control study.

BACKGROUND: A new technique exists that enables functional mapping of the retina. A control population was examined to obtain normative values and to assess the reproducibility of this new test. METHODS: Twenty healthy volunteers were tested using a 61 hexagonal array stimulus with a 14 minute recording period. Median 5th and 95th percentiles were determined for implicit times and amplitude measures for the 61 test areas. Repeat measurements were performed on 10 individuals. Wilcoxon and Bland and Altman techniques were used to quantify reproducibility of the test. RESULTS: The implicit time of the wave-form components was not found to vary over the retina (peak or b-wave component, 35.52 (1.4) ms; trough or a-wave component, 17.76 (0.8) ms). Reproducibility was found to decrease with eccentricity (coefficient of repeatability 17.4% for the central area increasing to 30.3% for the peripheral ring). CONCLUSIONS: The findings suggest that reproducibility, although variable with eccentricity, is comparable with conventional electrophysiology. These limits of variation were used to assign confidence intervals to individual retinal areas and will be used (future work) in the examination of diseased states.

Factorial phase imaging: a new concept in the analysis of first-pass cardiac studies.

A method is described which offers an alternative to conventional Fourier phase/amplitude imaging for analysing representative LV cycles obtained from first-pass cardiac studies. The method involves applying factor analysis to the data and producing functional images of the first two factor loadings after polar transformation. The method corrects for cycle asymmetries which produce anomalies in the Fourier images. For 88 studies the method provides greater uniformity among normal phase images while retaining, or even enhancing, abnormalities. Small differences between amplitude images produced by the two methods do not appear to be important.

A 195Hgm-195Aum generator for use in first-pass nuclear angiocardiography.

The development of a mercury-195m/gold-195m radionuclide generator system is described, together with the results of preliminary experiments in animals and in man. The daughter radionuclide has a half-life of 30.5 s with a principal gamma emission at 262 keV, ideal for use with the multicrystal camera but suitable also in conjunction with the Anger design. The half-life of the parent radionuclide is 40 h. The eluant in current use is 2.5 mM sodium cyanide, the column being stored under 2% sodium nitrate solution when not in use to prevent radiation damage. Generators capable of delivering 550 MBq of 195Aum in a 0.4 ml bolus have been used for first-pass nuclear angiocardiography, the half-life being long enough to permit administration via peripheral intravenous injection. The radiation dose associated with this radionuclide is much lower than that from conventional 99Tcm-labelled radiopharmaceuticals, permitting multiple sequential studies, even on children.

Mobile gamma cameras and 99mTc-labelled phosphates in acute myocardial infarction.

99mTc-labelled phosphates (PyP, IDP and MDP) were compared for their usefulness for the diagnosis of myocardial infarction. The results indicated the superiority of 99mTc-labelled imidodiphosphonate over the other 99mTc-labelled phosphates. At the same time, two mobile gamma cameras were compared with respect to their instrumental characteristics and clinical versalitity and their advantages and disadvantages are reported.

High-quality, high-throughput, early whole-body bone scanning.

99mTc-IDP allows for high-quality, high-throughput and early whole-body bone scanning. Whole-body bone scans obtained at 30 min after intravenous administration of the tracer were compared with the standard and delayed 180 min p.i images. A subtraction technique allowed for the anlysis of the 30 min scans and it was found that in both types of investigation the same clinical information was present. This procedure will therefore increase considerably the throughput of whole-body bone scans in a busy department of nuclear medicine. It is however inadequate for detailed analysis of the skeleton (evaluation of benign bone disease).

UK nuclear medicine survey, 1989/90.

A postal survey of UK nuclear medicine departments was carried out to obtain information on activity during the year 1989/90. A rise of 14% in the number of administrations of radiopharmaceuticals was found compared to 1982: a rise of 22% in imaging studies was offset by a 30% decrease in the number of nonimaging investigations. The estimated total number of administrations in the UK was 430,000.

Automated radiopharmaceutical dispensing.

An automated workstation, developed for clinical biochemistry, has been evaluated for potential use in preparing radiopharmaceuticals. It was found that substantial modifications, both hardware and software, would be required but the approach seemed feasible.

UK nuclear medicine survey, 1992-93.

A postal survey of UK nuclear medicine departments was undertaken to collate information on equipment, numbers of procedures and staffing levels for the years 1992 and 1993. It was estimated that there are 235 sites undertaking nuclear medicine, the total number of procedures performed being some 490,000 in 1993 compared with 430,000 in 1989. Informal investigation suggests that the increase is due to greater usage of myocardial perfusion and lung ventilation/perfusion studies. Wide variations were noted in staffing levels, with only 22% of departments having medical cover of half-time equivalent or better: over 30% of departments have less than one consultant session per week. Approximately 20% of departments claimed to have no physics input, with a further 20% having less than one session per week.

Investigation of factors affecting adhesion of 99Tcm labelled colloids to glass vials.

Factors which may possibly influence the adhesion of 99Tcm labelled colloid to glass vials were investigated. Of five colloids studied, two protein-free tin colloids were affected most by the problem, some 12-18% being lost by adhesion to the glass vial when the colloid was left resting on the bench, or 60-87% when it was rotated for a period of 4 h. The proportion of activity which adhered to the vial was dependent on the time the product had been contained within it, whether on the bench or rotated. Two of the colloids, both of which contained protein, were affected only slightly by the problem, one being a tin colloid and the other a sulphur colloid. An antimony sulphide colloid, which contained no added protein, was virtually unaffected by the problem. The degree of adhesion was not related to the pH of the product, nor to the temperature at which the product was stored, nor to the tendency of the colloid particles to grow in size upon vigorous agitation.

Formation of large particles in a 99Tcm-tin colloid preparation.

The effect of prolonged continuous movement on particle size within a 99Tcm-tin colloid preparation was investigated, making use of photon correlation spectroscopy for size estimation. No increase in particle size was found when the preparation was left undisturbed for up to 7 h, the mean (+/- S.D.) size being found to be 239 +/- 24 nm. When the preparation was placed in a laboratory rotator for the same period of time, there was a gradual increase in mean particle size to 404 +/- 127 nm. The increase in size was much greater, however, after 7 h of continuous agitation in a laboratory shaker, a mean value of 2260 +/- 746 nm being observed. An antimony sulphide colloid was subjected to continuous agitation for the same period of time but there was no evidence of aggregation. It is supposed that continuous movement of reconstituted 99Tcm-tin colloid, leading to a marked increase in particle size, may explain occasional instances of significant lung intake.

Formation of labelled colloid in 99Tcm-DMSA due to the presence of bactericidal fluid.

The possibility that traces of bactericidal fluid in 99Tcm-DMSA could lead to the formation of labelled colloid, was explored. In vitro investigations were undertaken using ultracentrifugation techniques and photon correlation spectroscopy. The latter showed that both contaminated and uncontaminated DMSA contained colloidal (or particulate) material. However the presence of 10 microliters bactericidal fluid as contaminant was shown by ultracentrifugation to result in labelling of this colloidal material when 99Tcm was added to DMSA. Studies in a normal volunteer confirmed the results of the in vitro studies, in that significant liver and spleen uptake was observed after the administration of contaminated 99Tcm-DMSA.

Technetium-99m-tetrofosmin: retention of nitrogen atmosphere in kit vial as a cause of poor quality material.

Technetium-99m (99Tc(m))-tetrofosmin was prepared using four different reconstitution methods. The radiochemical purity (RCP) of these products was assessed 8 h later using thin layer chromatography (TLC). Material produced using the original method supplied by the manufacturer and using an newer method, which involves the use of a vent needle and the addition of air, had acceptable RCP (mean +/- SD 94.2 +/- 1.1% and 94.7 +/- 1.7%, respectively) and similar chromatograms. In addition, both products showed good clinical efficacy and exhibited normal biodistribution behaviour. Preparing 99Tc(m)-tetrofosmin using the two other methods, one using a high radioactive concentration and the other maintaining the nitrogen content of the kit vial, gave rise to chromatograms with reduced RCP (63.5 +/- 10.9% and 61.9 +/- 7.6%, respectively) and greater levels of impurities. Although neither of these last two preparations was used clinically, we suggest that reports of poor quality images may be the result of administration of materials similar to these. Results for the high radioactive concentration method were as expected and are consistent with the restrictions imposed by the manufacturer. However, results using the last method are surprising and would suggest that the production of good quality 99Tc(m)-tetrofosmin is dependent on the quantity of nitrogen in the kit vial. We believe that the amount of nitrogen removed from the kit vial during the process of reconstitution is critical. If too much nitrogen is present this will result in poor quality material. In practice it is conceivable that there could be occasions when insufficient nitrogen is removed when following the manufacturer's original guidance, thereby leading to low RCP material. To ensure adequate nitrogen is removed during reconstitution, adoption of the manufacturer's revised method, involving the deliberate introduction of air, is therefore appropriate.

A study of doses to the hands during dispensing of radiopharmaceuticals.

Radiation doses to both the fingertip and finger base of staff preparing radiopharmaceuticals in a radionuclide dispensary have been monitored for more than 5 years. The records show a fall in dose thought to be due, largely, to the introduction of all-glass (lead) syringe shields. In 1998, the annual radiation dose (mean +/- standard deviation) to the fingertip had fallen to 83+/-11 mSv, which corresponded to 0.18 mSv per 10 GBq 99Tc(m) handled. There was less reduction in dose to the finger base, which, in 1998, was 70% of that to the fingertip. To study the effect of a syringe shield on the distribution of dose across the hands, model hands were constructed and a gamma extremity monitoring system (GEMS) was used to measure dose. When a radiopharmaceutical is dispensed, contributions to the dose from activity in the vial and syringe change during the procedure. As a result, relative doses to different fingers and the ratios of doses to the tip and base of each finger will fluctuate throughout dispensing. In the absence of a syringe shield, the front tip of the index finger received the greatest radiation dose. When a syringe shield was used, doses were lower and the dose distribution was completely different. These findings have important implications in terms of compliance with the new UK Ionizing Radiations Regulations, where the dose limit is applied to 'the dose averaged over any area of 1 cm2 regardless of the area exposed'.

The design and construction of a central radiopharmacy.

Following an inspection by the Medicines Control Agency (MCA) of the existing radiopharmacy, it was necessary to design and build a new facility at the Western Infirmary to supply radiopharmaceuticals for patient use throughout the West of Scotland. This is one of the largest radiopharmacies in Europe, supplying some 40,000 patient doses per year prepared using both closed and open procedures. The design parameters and construction details of the new facility are presented.

Immunoscintigraphy of human ovarian cancer xenografts using a radiolabelled monoclonal antibody to human pregnancy serum-derived immune complexes.

The in vivo imaging of xenografted human ovarian cancer in nude mice with a specific and control radiolabelled monoclonal antibody (MoAb) is described. The specific MoAb was previously raised by immunizing mice with immune complexes derived from late human pregnancy serum. In the first group of mice the specific MoAb 131I-5E3 F(ab')2 was injected, while a second group received equivalent amounts of a control MoAb 131I-UJ13A F(ab')2. The mice were imaged at various times up to a maximum of 2 weeks using a gamma camera, and the tumour to non-tumour (T/NT) ratio was recorded for each group. The T/NT ratio rose to 2.02 in the specific group, while the corresponding ratio in the control group was 0.70. In addition, the count rate in the tumour and non-tumour regions was determined on each imaging occasion. Biological half-lives of the divalent fragments of 5E3 and UJ13A in the tumour were 7.53 days and 0.62 days, respectively. Following sacrifice, the tumours were excised and counted relative to the rest of the animal, and the T/NT ratio was calculated. In vitro results were in direct agreement with those recorded in vivo using the gamma camera. From the results it would appear that the divalent fragment of 5E3, which has been raised to immune complexes derived from late human pregnancy serum, is specific for human ovarian tumour xenografts in the nude mouse model.