Relationship between sharpness and noise in CT images reconstructed with different kernels.

The sharpness expressed as the full-width at half-maximum of the line spread function (LSF) was derived for different CT scanners and reconstruction kernels. The LSFs were calculated from images of a 0.28 mm tungsten bead and successfully fitted to cumulative Gaussians by means of probit plots. The corresponding image noise was measured in a homogeneous material. The investigation showed that the noise increases strongly for small improvements in sharpness. The method was reproducible and insensitive to small positional shifts of the phantom.

A phantom for investigation of tumour signal and noise in PET reconstruction with various smoothing filters: experiments and comparisons with simulated intensity diffusion.

An abdomen phantom, containing an idealised fillable liver volume was developed. A fast settling gel was used to produce a row of simulated tumours as well as normal surrounding tissue, each of these components containing different concentrations of (18)F-FDG. The magnitude of the signal above normal tissue intensity as well as the noise was obtained after reconstructions with Ordered Subset expectation-maximization (OSEM) and various Gaussian smoothing filters. Signals derived by 'simulated intensity diffusion' agreed well with experimental results.

Preparation of a radionuclide/gel formulation for localised radiotherapy to a wide range of organs and tissues.

Localised radiotherapy by instillation of radiolabeled particles is being used to treat rheumatoid arthritis and certain tumors. Such therapy is limited to organs and tissues capable of retaining the radioactive compound until the radioactivity is sufficiently low, and the leakage to other parts of the body is no longer unacceptable. In this study, radiolabeled particles, i.e. 90Y-silica colloid particles, were encapsulated in calcium alginate gels, and the leakage of radioactivity from the gels was monitored. The purpose of the study was to develop a formulation suitable for the localised delivery of radiation therapy to a wide range of organs and tissues. An injectable gel formulation was developed, liberating only small amounts of radioactivity into the surrounding medium. The formulation is a viscous liquid at room temperature and forms a gel on heating to normal body temperature. Thus, it should be suitable for the localised delivery of radiolabeled particles to a wide range of organs and tissues. The study also includes a formulation exhibiting time-dependent gelation. However, this formulation was not found to be suitable for the purpose.

Intratumour heterogeneity in the uptake of macromolecular therapeutic agents in human melanoma xenografts.

Intratumour heterogeneity in the uptake of blood-borne technetium-labelled human serum albumin ((99m)Tc-HSA) was studied in human melanoma xenografts in an attempt to identify transport barriers leading to inadequate and heterogeneous uptake of macromolecular therapeutic agents in tumours. The Bioscope imaging system, which can detect the distribution of (99m)Tc in 10-microm-thick tissue sections with a spatial resolution of just above 50 microm, was used to image the (99m)Tc-HSA uptake. Xenografted tumours of four human melanoma cell lines were included in the study. Significant intratumour heterogeneity in the uptake of (99m)Tc-HSA was detected. The heterogeneity had two distinctly different components, one random and one radial component. The uptake was lowest in the centre of the tumours and increased towards the tumour periphery. This radial heterogeneity was superimposed by a random heterogeneity, that is, spots with high uptake colocalised with spots with high vascular density and regions without significant uptake colocalised with necrotic regions. The magnitude of the heterogeneity did not change significantly with time after the administration of (99m)Tc-HSA. The tumours showed a random and a radial heterogeneity in blood perfusion similar to that in the uptake of (99m)Tc-HSA. The observations reported here suggest that the intratumour heterogeneity in the distribution of (99m)Tc-HSA was initiated primarily because of heterogeneity in the supply of (99m)Tc-HSA through the microvasculature, and that the presence of severe transport barriers in the tumour interstitium prevented significant equalisation of the initial heterogeneity with time. Consequently, strategies for improving the delivery of macromolecular therapeutic agents to tumours should focus on increasing the tumour blood perfusion to increase the total uptake and improving the diffusion conditions in the tumour interstitium to diminish the heterogeneity in the uptake.

Radionuclide therapy with bone-seeking compounds: Monte Carlo calculations of dose-volume histograms for bone marrow in trabecular bone.

The purpose of the present work was to investigate how haematopoietic stem cell survival is affected by the differences in the dose distribution that arise from different radionuclides contained in bone-seeking radiopharmaceuticals. This was carried out in three steps: (a) calculations of representative dose distributions in individual bone marrow cavities that are irradiated by sources of 89Sr, 186Re, 117mSn or 153Sm, uniformly distributed on the bone surfaces; (b) assessment of the corresponding haematopoietic stem cell survival and (c) a comparison of these results with results obtained using the assumption of a uniform dose distribution. Two different idealized models of the geometry of trabecular bone were formulated, each consisting of an infinite array of identical elements. Monte Carlo simulations were used to generate dose-volume histograms that were used to assess haematopoietic stem cell survival with two different assumptions about spatial cell distributions. Compared with a homogeneous dose distribution, the estimated cell survival was markedly higher for 117mSn and 153Sm, and only slightly different for 89Sr and 186Re. The quantitative results differed between the two geometric models and the assumptions about spatial cell distribution, but the trends were the same. The results imply that it is necessary to include dose distributions for individual bone marrow cavities in considerations concerning bone marrow toxicity.

Dose distribution measurements by MRI of a phantom containing lung tissue equivalent compartments made of ferrous sulphate gel.

Gel dosimetry based on magnetic resonance imaging (MRI) has previously been shown to provide verification of calculated dose distributions in soft tissue equivalent homogeneous phantoms. This study demonstrates how measurements of dose distribution can also be achieved in a phantom containing porous, lung-equivalent, Fricke gel. A phantom was made of Fe2+ infused low-density gel and conventional ferrous sulphate gel, filled in separate compartments in a Perspex container. Absorbed dose measurements were accomplished by MR imaging and by calibrating the proton spin-lattice relaxation rate (R1) versus absorbed dose by means of TLD measurements. This study shows that the production of lung-equivalent low-density (LD) dosimeter gel (mean CT number of -610 HU) is feasible. The MR signal detected in the LD gel dosimeter was substantially more noisy (i.e. displayed larger random fluctuations) than the signal from conventional gel, as expected. A deviation between calculated (TPS) and measured dose of about 3% (6 MV) and 4-7% (15 MV) was found in the LD region of the phantom. These results correspond well with data from other studies of dose distribution in lung-equivalent phantoms. The Fe2+ infused LD gel therefore seems suitable for measurement of absorbed dose distribution in phantoms that contain lung tissue compartments.

Methods for separation of contributions from two radionuclides in autoradiography with a silicon strip detector.

Two methods have been developed for the separation of contributions from two different nuclides in an autoradiography picture. Both approaches, a modified least squares (LS) and a maximum likelihood (ML) algorithm, are based on the position and energy information available from a digital detector. Tests and comparisons, using artificially as well as measured data, demonstrate that the ML approach performs slightly better, but is much more computationally demanding than the LS method.

Measurement of the extracellular volume of human melanoma xenografts by contrast enhanced magnetic resonance imaging.

The magnitude of the extracellular volume fraction (ECV) of tumors is of importance for the transport of macromolecular therapeutic agents from the vessel wall to the tumor cells. The aim of this study was to develop a method for measurement of tumor ECV by contrast enhanced MRI. Tumors of two human amelanotic melanoma xenograft lines (A-07 and R-18) grown intradermally in Balb/c nu/nu mice were used as model system, and muscle tissue was used as control. The renal arteries of the mice were ligated prior to i.v. administration of Gd-DTPA, and an MRI protocol for calculating Gd-DTPA concentration in tissue was followed. ECV was calculated from the Gd-DTPA concentrations in the tissue and in a plasma sample. In muscle tissue, the concentration reached a constant level after 1 min and the ECV was calculated to be 0.12 (+/- 0.01), consistent with values reported in the literature. Individual tumors showed large differences in the uptake of Gd-DTPA. The Gd-DTPA concentration in the tissue at 40 min after the Gd-DTPA administration was used to calculate tumor ECV. The ECV was found to differ significantly among regions of individual tumors and among individual tumors. The ECV ranged from 0.075 to 0.33 for A-07 tumors and from 0.016 to 0.097 for R-18 tumors. The intra- and intertumor heterogeneity in ECV was confirmed by histologic findings, showing that contrast enhanced MRI is suitable for non-invasive studies of the ECV in experimental tumors without necrosis.

Internal radionuclide therapy of primary osteosarcoma in dogs, using 153Sm-ethylene-diamino-tetramethylene-phosphonate (EDTMP).

Fifteen dogs were referred because of a spontaneous bone tumor, lameness, and local pain. The osteosarcoma diagnosis was established by clinical examination, X-ray, bone scintigraphy, and histological examination of biopsy material. The tumors were located in the extremities (n = 12), scapula (n = 1), maxilla (n = 1), and the frontal bone (n = 1). The dogs were given one to four i.v. injections of 153Sm-labeled ethylene-diamino-tetramethylene-phosphonate (153Sm-EDTMP; 36-57 MBq/kg body weight). Three dogs had surgery in addition to the radionuclide treatment. Platelet and WBC counts showed a moderate and transient decrease. No other toxicity was observed. Average tumor doses after a single injection were approximately 20 Gy, considerably higher in some areas because of inhomogeneous uptake. Macroscopically distant metastases were detected in seven dogs at autopsy. One dog died from an intercurrent disease, free of cancer, 5 months after the radionuclide treatment. None of the dogs was cured. The median and mean survival times from the first treatment to death or euthanasia were 150 and 252 days, respectively. Nine of the dogs had obvious pain relief, and five of them seemed pain-free: one for 20 months and one for 48 months. It is concluded that high tumor doses may be deposited in dog osteosarcomas by 153Sm-EDTMP, and the ratio between tumor dose and the dose to surrounding tissues is favorable. The treatment gives pain relief and in some cases tumor growth delay. In combination with surgery, 153Sm-EDTMP may prolong life significantly and possibly cure the disease because the development of metastases are seemingly postponed. No serious side effects were observed.

An anthropomorphic phantom for receiver operating characteristic studies in CT imaging of liver lesions.

The purpose of the study was to develop a methodology that allowed quantitative assessment of image quality in CT and its relationship to dose. An anthropomorphic phantom was designed for use in receiver operating characteristic (ROC) studies of the detectability of liver lesions with CT. The lesions were simulated by different mixtures of glycerol and water that were filled into holes of different diameters in a liver tissue substitute. A pilot study was carried out on five different scanners that were operated at various exposure settings. A positive correlation was demonstrated for each of the scanners between the weighted CT dose index (CTDIW) and the area under the ROC curve. For the exposure settings used in the clinical routine in the five laboratories, the CTDIW ranged from 15 to 31 mGy. Three observers who read the corresponding set of five phantom images agreed, as judged from the areas under the ROC curves, that there was a marked difference in quality between the three best images and the other two. The two newest scanners in the study had the lowest CTDIW, and at the same time the best ROC results. The phantom and the ROC methodology may, with a set of suggested improvements, be used for comparison of the performance in different CT laboratories, and to establish the dose needed to ensure adequate image quality for a particular scanner.

[Measurement of glomerular filtration rate in connection with 99Tc-DTPA renography]. / Måling av glomerulaer filtrasjonshastighet i tilknytning til renografi med 99mTc-DTPA.

Glomerular filtration rate (GFR) was measured in two groups of cancer patients. In 20 patients, glomerular filtration rate was measured simultaneously with 51Cr-EDTA, 99mTc-DTPA and the X-ray contrast agent iohexol as markers, and with a complete set of eight blood samples during 24 hours. In a second group of 120 patients, we used 99mTc-DTPA only and tested various simplified methods, based on one or two blood samples. Glomerular filtration rate was also calculated from serum creatinine. There was excellent agreement between the values measured with the three markers, and, in the same group of patients, very good agreement with the results of simplified methods. The larger study carried out with 99mTc-DTPA on the second group of patients, confirmed a very good agreement between methods based on the slope of the plasma curve and a method based on one blood sample only. The correlation was worse between the values obtained by any of the radionuclide methods and those calculated from serum creatinine. The latter method should therefore not be used for determination of renal function.

Methods for automatic detection of peristaltic waves in the small intestine.

Methods have been developed to detect peristaltic waves that carry radioactive technetium-99m mebrofenin excreted in the bile through the upper intestinal tract. A dynamic study is processed sequentially to determine the region of interest in the intestine, ROI1, where the activity appears earliest. Mathematical correlation is calculated between the time-activity curves for all pixels exceeding a threshold, excluding those belonging to the liver and ROI1, and the time-activity curve for ROI1. Pixels with maximum correlation at the same time shift are assumed to belong to the same region of the intestine. Thus, for increasing time shifts, a consecutive series of ROIs are formed. The computer then builds an image (peristaltogram) where each line contains the counts versus time variation for an ROI, and the sequence of lines corresponds to the sequence of ROIs. Based on template matching, the periods of the peristaltic contractions in the small intestine may be determined. A case study demonstrates that peristaltic waves easily may be recognized and characterized.

Detectability of simulated masses and calcifications in mammography. Development of a phantom and a new method for determination of receiver operating characteristics.

PURPOSE: To develop a method for receiver operating characteristics (ROC) studies in mammography. MATERIAL AND METHODS: We developed a phantom based on excised breast tissue and overlay tiles that could be arranged in an arbitrary pattern across the surface of the breast tissue. Some of the tiles contained structures simulating calcifications or masses that produced image contrast near the experimentally determined detection threshold. Based on this phantom, a methodology for performing ROC studies in mammography was developed. The ROC curves were constructed from reporting schemes filled in by radiologists at five different laboratories. The curves were determined by a novel method: a non-linear least-squares fit of a mathematical model to the data. RESULTS: There were large differences among the areas under the ROC curves obtained from the five laboratories.

Assessment of image quality and total performance in Norwegian mammography laboratories. Findings in a national survey based on different phantoms and ROC methodology.

PURPOSE: To assess the image quality at different mammography laboratories. MATERIAL AND METHODS: Two commercial mammographic test phantoms and one phantom based on excised mammary tissue were used in an assessment of the imaging chain and total performance at 45 Norwegian mammography laboratories. The breast-tissue phantom was used for a receiver operating characteristics (ROC) analysis. This was carried out by putting overlays with identifiable regions (some of which contained a cluster of simulated calcifications) on top of the mammary tissue, and then having a radiologist report the confidence of a finding for each region. RESULTS AND CONCLUSION: The areas under the ROC curves were in general high. In nearly all the laboratories, performance was improved when a magnification technique was applied. There were wide variations among the laboratories in total performance as measured by the area under the ROC curve, and also in the physical parameters derived by means of the commercial phantoms. In general, a good ROC performance was associated with a good physical performance in the imaging chain.

Measurement of perfusion rate in human melanoma xenografts by contrast-enhanced magnetic resonance imaging.

Reliable methods based on MRI for measurement of the perfusion rate in human tumors are highly warranted. Tumors of two amelanotic human melanoma xenograft lines were subjected to dynamic 1H MRI after i.v. administration of gadopentetate dimeglumine (Gd-DTPA). The aim was to investigate to what extent different perfusion parameters determined from the Gd-DTPA kinetics, i.e., the initial uptake rate, the maximal uptake, the decay rate, and the perfusion rate calculated by use of the Kety equation, can be used as a reliable estimate of tumor perfusion rate. Each parameter was calculated in dual; one calculation was based on relative signal intensity increase (RSII) in T1-weighted MR images and the other on Gd-DTPA concentration determined from the images. The perfusion parameters were compared with the perfusion rates determined from measurement of tumor uptake of 86Rb or [14C]iodoantipyrine. The results showed that reliable estimates of tumor perfusion rate can be achieved from analysis of Gd-DTPA kinetics by use of the Kety equation. Gd-DTPA kinetics based on concentration might be used to achieve reliable estimates of absolute tumor perfusion rate, whereas reliable estimates of the relative perfusion rate might also be achieved from Gd-DTPA kinetics based on RSII. The initial uptake rate, the maximal uptake, and the decay rate of Gd-DTPA, however, are not reliable estimates of tumor perfusion rate, mainly because these parameters are highly influenced by the tumor extracellular volume fraction in addition to the perfusion rate.

Predicting detectability in radioguided surgery: a study based on physical characterization of the probe and on tissue uptake ratios of 125I-, 153Sm- and 165Er-labelled radiopharmaceuticals.

We performed measurements of the sensitivity in water of the radioguided surgery system, Neoprobe 1000. Point sources of 125I, 153Sm and 165Er were measured to a geometric accuracy of 0.1 mm. These measurements were performed with the detector uncollimated and with two different collimators; the results were used to construct three-dimensional sensitivity matrices from which isosensitivity curves could be derived. The matrices were used to simulate the reliability of detecting tumours of various sizes, at various depths, and with variable uptake ratios and activity levels of the three radionuclides investigated. Simulations showed improved tumour detection with a conical collimator, demonstrating that collimator design is important. To minimize the radiation dose during diagnostic work-up, a novel bone-seeking radiopharmaceutical, 165Er-EDTMP, was developed and tested. It showed rapid bone localization, especially in growth zones, and rapid clearance from blood and non-osseous tissues. Its biodistribution was similar to that of 153Sm-EDTMP. Based on these results and previous clinical data, the simulations predicted that lung metastases with a radius of 1 mm might be reliably detected with the Neoprobe 1000 system, if equipped with a suitable collimator.

Measurement of the ferric diffusion coefficient in agarose and gelatine gels by utilization of the evolution of a radiation induced edge as reflected in relaxation rate images.

A method has been developed to determine the diffusion coefficients of ferric ions in ferrous sulphate doped gels. A radiation induced edge was created in the gel, and two spin-echo sequences were used to acquire a pair of images of the gel at different points of time. For each of these image pairs, a longitudinal relaxation rate image was derived. From profiles through these images, the standard deviations of the Gaussian functions that characterize diffusion were determined. These data provided the basis for the determination of the ferric diffusion coefficients by two different methods. Simulations indicate that the use of single spin-echo images in this procedure may in some cases lead to a significant underestimation of the diffusion coefficient. The technique was applied to different agarose and gelatine gels that were prepared, irradiated and imaged simultaneously. The results indicate that the diffusion coefficient is lower in a gelatine gel than in an agarose gel. Addition of xylenol orange to a gelatine gel lowers the diffusion coefficient from 1.45 to 0.81 mm2 h-1, at the cost of significantly lower Rl sensitivity. The addition of benzoic acid to the latter gel did not increase the Rl sensitivity.

Determination of 3D dose distribution from intracavitary brachytherapy of cervical cancer by MRI of irradiated ferrous sulphate gel.

BACKGROUND AND PURPOSE: MRI ferrous sulphate gel dosimetry has proven to be a valuable method for assessment of dose delivered in teletherapy. The intention of this study was to investigate ferrous sulphate gel as a possible dosimeter for intracavitary brachytherapy applications. MATERIALS AND METHODS: A plastic duplicate of a cervix ring applicator set was submerged in Fe2(+)-infused gelatin gel. The gel was subsequently irradiated by a stepwise moving 192Ir source, using automatic afterloading equipment (Microselectron, Nucletron-Oldelft International BV, Veenendaal, The Netherlands). A 3D dose distribution was reconstructed from MR images of the gel. RESULTS: The gel dose measurements were found to be of the same accuracy as TLD measurements. Isodose curves based on gel dosimetry and isodose curves computed by a dose planning system were generally less than 2 mm apart. MR images showing the position of the applicator set in a patient during treatment were used to obtain images describing patient anatomy in the sagittal and ring planes of the applicator set. Isodose curves computed from the gel measurements were then superimposed on these images, illustrating one possible way of linking dosimetrical and anatomical data. CONCLUSIONS: Our study shows that MRI ferrous sulphate gel dosimetry is a useful tool for studies of dose distributions in brachytherapy and their relation to critical organs. Possible improvements of the gel dosimeter lie in reducing the diffusion of ferric ions and increasing the radiation sensitivity of the gel.

Proton relaxation times and interstitial fluid pressure in human melanoma xenografts.

The interstitial fluid pressure (IFP) and the proton spin-lattice and spin-spin relaxation times (T1 and T2) of some experimental tumours have been shown to be related to tumour water content. These observations have led to the hypothesis that magnetic resonance imaging (MRI) might be a clinically useful non-invasive method for assessment of tumour IFP. The purpose of the work reported here was to examine the general validity of this hypothesis. R-18 human melanoma xenografts grown intradermally in Balb/c nu/nu mice were used as the tumour model system. Median T1 and T2 were determined by spin-echo MRI using a 1.5-T clinical whole-body tomograph. IFP was measured using the wick-in-needle technique. No correlation was found between tumour IFP and fractional tumour water content. Moreover, there was no correlation between median T1 or T2 and IFP, suggesting that proton T1 and T2 values determined by MRI cannot be used clinically to assess tumour IFP and thereby to predict the uptake of macromolecular therapeutic agents.

Measurement of dynamic wedge angles and beam profiles by means of MRI ferrous sulphate gel dosimetry.

The purpose of this study is to examine the possible value of measuring the dose distribution in dynamic wedge photon beams using ferrous sulphate gel phantoms analysed by MRI. The wedge angles and dose profiles were measured for a field size of 70 x 70 mm2 and for dynamic wedge angles of 15 degrees, 30 degrees, 45 degrees and 60 degrees using a 15 MV photon beam generated from a Clinac 2100 CD (Varian). The dose profiles obtained from MRI ferrous sulphate gel were in good agreement with the dose measurements performed with a diode detector array. Also, the wedge angles determined from the MRI ferrous sulphate gel agreed well with the values obtained by using film dosimetry and with calculations by use of TMS (treatment planning system) (Helax, Uppsala, Sweden). The study demonstrated that MRI ferrous sulphate gel dosimetry is an adequate tool for measurements of some beam characteristics of dynamic radiation fields.