- Email: [email protected]
Sub-Millimeter Technetium-99m Calibration Sources
PII S1536-1632(02)00084-7
Molecular Imaging and Biology Vol. 4, No. 5, 380–384. 2002 Copyright © 2002 Elsevier Science Inc. Printed in the USA. All rights reserved. 1536-1632/02 $–see front matter
ARTICLE
Sub-Millimeter Technetium-99m Calibration Sources Jeffrey R. English, CNMT1, Roberto Accorsi, PhD2, John D. Idoine, PhD3, J. Anthony Parker, MD, PhD1, Jürgen T. Renze, PhD4, Richard C. Lanza, PhD5, John V. Frangioni, MD, PhD4 1
Division of Nuclear Medicine, Department of Radiology, Beth Israel Deaconess Medical Center, Boston, MA, 2Department of Radiology, The Children’s Hospital of Philadelphia, Philadelphia, PA, 3Department of Physics, Kenyon College, Gambier, OH, 4Division of Hematology/Oncology, Department of Medicine, Beth Israel Deaconess Medical Center, Boston, MA, 5Department of Nuclear Engineering, Massachusetts Institute of Technology, Cambridge, MA Purpose: Small animal radioscintigraphic imaging systems aim to achieve sub-millimeter resolution. At the present time, sub-millimeter calibration sources that can be placed at will within an imaged volume are not readily available. We have developed a method for producing technetium-99m (Tc-99m) sources in less than 15 minutes with readily available reagents. Procedures: Tc-99m pertechnetate [TcO4] was incubated with 45 m to 106 m diameter spherical anion exchange beads, washed, and mounted as desired for instrument calibration. Results: The procedure yields spherical sources having between 6.8 Ci to 11.1 Ci of Tc99m per source. This work shows that dual imaging of these sources using white light and radioscintigraphy permits measurement of system performance with high precision. Conclusion: Easily prepared, sub-millimeter Tc-99m spherical calibration sources are described, and it is demonstrated that such sources are useful for measuring the resolution and sensitivity of radioscintigraphic systems, such as those designed for small animal imaging. (Mol Imag Biol 2002;4:380–384) © 2002 Elsevier Science Inc. All rights reserved. Key Words: Radioscintigraphy; Single-photon emission computed tomography (SPECT); Technetium-99m; Calibration; Resolution; Coded aperture imaging.
Introduction
R
adioscintigraphy plays an important role in cancer detection and staging. A major focus of research in this area is the development of tumortargeted radiopharmaceuticals. This in turn requires preclinical testing in small animal models. Although clinical single photon emission computer tomography (SPECT) imaging systems typically have a maximum resolution of approximately 5–8 mm, newer small animal SPECT systems are reported to have resolutions approaching 830 m.1–3 Other high-resolution techniques, such as those employing pinholes, micro-collimators, and Compton cameras are also being developed. As a consequence, there is a general need for calibration sources, of the radioisotope under study that are far smaller than imaging system resolution and are of a known geometry. Address correspondence to: John V. Frangioni, M.D., Ph.D. HIM1023, Beth Israel Deaconess Medical Center 330 Brookline Avenue Boston, MA 02215. E-mail: [email protected]
380
Since there are several problems associated with SPECT imaging of small animals, including the need for relatively high doses of isotope, long scan times, and limited resolution, alternatives have been developed. One such technique utilizes a coded aperture, that is, a radio-opaque mask that is placed between the camera crystal and the object.4 The mask has a pattern of pinholes of fixed diameter, whose pattern is such that hundreds or even thousands of overlapping projections of the object fall on the detector crystal. Using straightforward algorithms based on the hole arrangement, the image is reconstructed. For many common objects, particularly those that are bright objects on a dark field, coded aperture imaging provides higher sensitivity, and depending on the design of the system, higher resolution and/or shorter scan times than SPECT.4 Like SPECT and other high-resolution techniques, there is a need for sub-millimeter calibration sources. In vivo radioscintigraphic imaging can be accomplished using targeted ligands associated with one of many possible radioisotopes. Some of the most com-
Sub-Millimeter Tc-99m Calibration Sources/ English et al. 381
mon isotopes used pre-clinically and clinically are iodine-123, indium-111, and Tc-99m. Of these, Tc-99m has many favorable characteristics including 140 keV monoenergetic gamma emission, a six hour half-life, low radiation absorbed dose, availability from a generator, and low cost. Indeed, nearly 80% of all radiopharmaceuticals used in clinical nuclear medicine are Tc99m-labeled compounds, and various strategies exist for labeling antibodies, peptides, and small molecules with it.5–9 In this work, a simple method for producing Tc-99m calibration sources having a diameter sufficiently small to permit characterization of imaging systems with submillimeter resolution is presented. In addition, the utility of these sources using a coded aperture imaging system is proven.
Materials and Methods Reagents AG1–X8 anion exchange resin in its formate form (catalog #140-1454) and Poly-Prep disposable plastic columns (catalog #731–1550) were purchased from BioRad (Hercules, CA). Tc-99m pertechnetate was eluted from a molybdenum-99 generator using normal saline. Water was purified to 18 M on a Milli-Q apparatus (Millipore, Bedford, MA).
periment, 10 L to 50 L of this slurry was placed on a Superfrost glass slide (Fisher Scientific, Hanover Park, IL). Individual beads were positioned using a razor blade and the slide was blotted with a paper towel to remove unwanted beads. A slide loupe or magnifying glass is used to ensure that the desired number of beads is on the slide and at the desired location. Excess water is then blotted and the slides are allowed to dry. For activity per bead/slide of greater than 10 Ci, measurement was made using a model CRC-15R dose calibrator (Capintec, Ramsey, NJ). For lower activities, or whenever system sensitivity was being measured, a Packard Cobra II well detector (Perkin Elmer Life Sciences, Boston, MA) was used.
White Light Imaging of Calibration Sources For single bead experiments, or those where bead separation was less than 5 mm, brightfield images were acquired using a calibrated Nikon (Melville, NY) Eclipse TE-300 microscope fitted with a Photometrics Sensys CCD Camera (Model 1401, Roper Scientific, Tucson, AZ). For experiments where bead separation was greater than 5 mm, a previously described small animal imaging system was used in darkfield mode.10 Image acquisition and analysis was performed using IPLab software (Scanalytics, Fairfax, VA).
Coded Aperture Radioscintigraphy Preparation of Calibration Sources All procedures were performed behind a 0.5-inch lead shield with a 0.25-inch leaded glass window (Biodex, Shirley, NY). Approximately 50 mCi of Tc-99m pertechnetate in 1 ml to 3 ml of normal saline was placed in a Poly-Prep column with the tip still attached. Poly-Prep columns have a porous 30 m polyethylene bed support that retains the anion exchange beads. One hundred milligrams of AG1-X8 resin was placed in a 1.5 ml Eppendorf tube, washed three times and resuspended in water at a concentration of approximately 200 beads per L. Five microliters (approximately 1000 beads) of this slurry was added to the Poly-Prep column, and the solution was pipetted up and down for two minutes to achieve constant bead collision with the Tc-99m pertechnetate. The tip of the Poly-Prep column was removed and the flow-through was discarded. Beads were washed with 20 ml of water, and the column was capped with the supplied yellow cap. One milliliter of water was added to the column, the beads were brought into suspension by up/down pipetting, and transferred to a 5 ml polystyrene test tube. The volume was adjusted to 4 ml with water and the beads were pelleted by centrifugation at 500 g for 30 seconds. The beads are barely visible on the bottom of the tube. All but approximately 200 L of the supernatant was removed, and beads were resuspended by up/down pipetting. Depending on the ex-
Coded aperture radioscintigraphy using a conventional Anger camera has been described in detail previously.4 Briefly, the mask was constructed from 1 mm thick sintered tungsten alloy, Catalog #K1750, manufactured by H.C. Starck Kulite Tungsten Products (East Rutherford, NJ). One point one mm holes were drilled in a 62 by 62 no-two-holes-touching pattern using a modified uniformly redundant array.4 The mask was fitted on a Forte gamma camera (ADAC Laboratories, Milpitas, CA) using a custom aluminum frame such that the detector to mask distance was 30 cm, the mask to object distance was 10 cm, the magnification factor was 3.0 and the field of view was 9.3 cm. Five hundred thousand total counts were acquired in a 512 by 512 matrix. Reconstructions were performed using MatLab v. 6.1.0.450, release 12.1 (MathWorks, Natick, MA) running on a 550 MHz Pentium III processor.
Results Calibration Source Preparation and Uniformity AG1–X8 anion exchange resin is composed of spherical beads with wet diameters in the range 45 m to 106 m. Using the protocol described in Materials and Methods and shown in Figure 1, individual beads were loaded with Tc-99m pertechnetate. In four separate ex-
382 Molecular Imaging and Biology, Volume 4, Number 5
the length of time, and/or the amount of agitation of the initial binding reaction. The relatively low percent bound was acceptable to us, however, since one of the goals was to prepare calibration sources in as short a time as possible. The entire procedure took less than 15 minutes. It should be noted that the capacity of the beads (1.2 mEq per ml) was more than 1500 times higher than the number of moles actually loaded.
Single-Source Measurements
Figure 1. Calibration Source Preparation. The flowchart shows the steps taken, as described in detail in Materials and Methods, for preparing sub-millimeter calibration sources.
periments, the amount of activity per bead was between 6.8 Ci to 11.1 Ci. For a typical labeling of 1000 beads with 50 mCi of Tc-99m pertechnetate using the protocol described in Materials and Methods, 13.6% to 22.6% of total activity was bound to the beads. This number was easily manipulated by changing the number of beads,
Since the diameter of individual beads varied widely, beads to be used for size standards were adsorbed to glass slides and their diameter measured precisely using light microscopy (see Materials and Methods). After air drying, the strength of bead adsorption was such that beads did not move during transport, shaking, or light blowing. Figure 2 shows the results from a typical experiment. In this case, the single calibration bead had a measured diameter of 96.3 m. Coded aperture reconstruction of the image took approximately 812 msec and is also displayed in Figure 2. By projecting a line through the center of the bead, the system resolution, that is, the full-width half-maximum (FWHM) of intensity, could be measured. The theoretical system resolution, which is a function of the mask and camera characteristics, was 1.67 mm. The FWHM measured after coded aperture reconstruction of the single 96.3 m source was 1.70 0.37 mm (Figure 2). It is interesting to note that the theoretical system resolution of 1.67 mm was obtained from an advanced mathematical model of the point spread function of
Figure 2. Single-Source Instrument Calibration. A single anion exchange bead measuring 96.3 m was labeled with 11.1 Ci of Tc-99m and imaged using brightfield microscopy (left panel). The same bead was imaged with a coded aperture imaging system, and the tomographic reconstruction through the center of the bead was displayed (middle panel; bead at arrow). Intensity (arbitrary units) is shown as an insert. The horizontal line projection through the bead’s center (right panel) was used to determine system resolution (FWHM 1.70 0.37 mm).
Sub-Millimeter Tc-99m Calibration Sources/ English et al. 383
the system.11 Use of the classic formula, where the geometric (1.1 mm * 4 / 3 1.47 mm) and system (3.7 mm / 3 1.23 mm) resolution add in quadrature, gives a system resolution of 1.92 mm. Using a small point source was important for an experimental validation of the prediction of the more advanced model with no need to deconvolve the size and shape of the point source that are often not known with sufficient accuracy. Single source measurements are also quite useful for measuring system sensitivity. Prior to imaging, a glass slide with a single Tc-99m bead was counted as described in Materials and Methods. The same bead was then imaged as described for Figure 2 and its count rate recorded. One of the main advantages of coded aperture imaging is its increased sensitivity over methods such as SPECT. Indeed, a single bead loaded with 11.1 Ci of Tc-99m yielded count rates on our system of approximately 2200 cps (11,892 cpm/Ci), with the 500,000 total counts desired for image reconstruction being acquired in less than four minutes.
Multiple-Source Measurements Single Tc-99m labeled sources can be positioned anywhere in three dimensions. Shown in Figure 3 is a planar arrangement of three beads in an asymmetrical
triangle. Beads are labeled A, B, and C, with the distances between beads A-B, A-C, and B-C being 9.3 mm, 12.7 mm, and 11.3 mm, respectively, as measured by darkfield illumination. The distances measured using coded aperture imaging and reconstruction were 9.3 mm, 13.0 mm, and 11.2 mm. These are within the error associated with the size of an individual pixel ( 0.37 mm).
Discussion The goal of this work was to produce sub-millimeter Tc-99m sources for measuring the resolution and sensitivity of small animal imaging systems. The procedure requires less than 15 minutes to complete and uses readily available, relatively inexpensive reagents. Although the sources described in this study utilized Tc-99m pertechnetate, the procedure should be readily adaptable to other anionic radioisotopes. To be used with cationic radioisotopes, one need only to use a cation exchange resin in place of the anion exchange resin. The geometry of the beads used in this study is of particular importance. Despite being dried on a glass slide, the beads remain essentially spherical as evidenced by both light microscopy and coded aperture imaging in orthogonal planes (data not shown). For ultra-high resolution systems, spherical source geometry
Figure 3. Multiple-Source Instrument Calibration. Three anion exchange beads (arrows), each measuring between 88 m to 98 m, were loaded with approximately 11.1 Ci of Tc-99m per bead and imaged using darkfield illumination (left panel). A black circle was drawn around each bead on the slide with a marker. The coded aperture reconstruction of the three beads is shown on the right panel. Beads are labeled A, B, and C, as described in the text. Intensity (arbitrary units) is shown as an insert.
384 Molecular Imaging and Biology, Volume 4, Number 5
should permit detection of system inhomogeneities that might only be seen with a particular detector position. In this study, multiple beads were placed in a single plane (glass slide). The bead format, however, permits custom holders to be constructed that position the sources anywhere in three-dimensional space, thus permitting testing of complex geometries. Finally, although we imaged the calibration sources using a coded aperture system, they should perform similarly well in SPECT, and other types of radioscintigraphic imaging systems.
3.
4.
5. 6.
Conclusion We have developed a simple method for preparing submillimeter, spherical Tc-99m sources. Such sources may be quite useful for calibrating the next generation of high-resolution, high-sensitivity radioscintigraphic small animal imaging systems. We thank Alec M. DeGrand for technical assistance and Grisel Rivera for administrative assistance. RA and RCL were funded by the Office of National Drug Control Policy Contract #DAAD07-98-C-0117 and the Federal Aviation Administration Grant #93G-053. JTR was supported by a fellowship within the Postdoc-Programme of the German Academic Exchange Service (DAAD). JI and JVF were funded by grant #R21CA88870 from the National Institutes of Health.
7.
8. 9.
10.
References 1. Habraken, J.B.; de Bruin, K.; Shehata, M.; et al. Evaluation of high-resolution pinhole SPECT using a small rotating animal. J. Nucl. Med. 42:1863–1869; 2001. 2. Liu, Z.; Kastis, G.A.; Stevenson, G.D.; et al. Quantitative analysis of acute myocardial infarct in rat hearts with is-
11.
chemia-reperfusion using a high-resolution stationary SPECT system. J. Nucl. Med. 43:933–999; 2002. Acton, P.D.; Kung, M.P.; Hou, C.; Plossl, K.; Keeney, C.; Kung, H.F. Ultra-high resolution single photon emission tomography imaging of the mouse striatum. Eur. J. Nucl. Med. Mol. Imaging. 29:446; 2002. Accorsi, R.; Gasparini, F.; Lanza, R.C. A coded aperture for high-resolution nuclear medicine planar imaging with a conventional anger camera: experimental results. IEEE Trans. Nucl. Sci. 48:2411–2417; 2001. Dilworth, J.R.; Parrott, S.J. The biomedical chemistry of technetium and rhenium. Chem. Soc. Rev. 27:43–55; 1998. Mahmood, A.; Kuchma, M.H.; Freiberg, E.; Goldstone, J.; Davison, A.; Jones, A.G. Functionalized tetradentate N2S2 chelates and their technetium-99m and rhenium complexes: synthesis, spectroscopy and structural characterization. In: Nicolini, M., Mazzi, U., eds. Technetium, rhenium and other metals in chemistry and nuclear medicine. Padova, Italy: Servizi Grafici Editoriali; 1999:253–257. Mahmood, A.; Kuchma, M.H.; Goldstone, J.; Morse, C.; Davison, A.; Jones, A.G. An N2S2 tetradentate chelate for solid-phase synthesis: evaluation in solution and solid phase and characterization of technetium-99 complexes. In: Nicolini, M., Mazzi, U., eds. Technetium, rhenium and other metals in chemistry and nuclear medicine. Padova, Italy: Servizi Grafici Editoriali; 1999:71–76. Jurisson, S.S.; Lydon, J.D. Potential technetium small molecule radiopharmaceuticals. Chem. Rev. 99:2205–2218; 1999. Liu, S.; Edwards, D.S. 99mTc-labeled small peptides as diagnostic radiopharmaceuticals. Chem .Rev. 99:2235–2268; 1999. Zaheer, A.; Lenkinski, R.E.; Mahmood, A.; Jones, A.G.; Cantley, L.C.; Frangioni, J.V. In vivo near-infrared fluorescence imaging of osteoblastic activity. Nat Biotechnol. 19:1148–1154; 2001. Accorsi, R., Design of near-field coded aperture cameras for high-resolution medical and industrial gamma ray imaging. Ph.D. Thesis, Nuclear Engineering, Massachusetts Institute of Technology: Cambridge, MA. 2001.
Molecular Imaging and Biology Vol. 4, No. 5, 380–384. 2002 Copyright © 2002 Elsevier Science Inc. Printed in the USA. All rights reserved. 1536-1632/02 $–see front matter
ARTICLE
Sub-Millimeter Technetium-99m Calibration Sources Jeffrey R. English, CNMT1, Roberto Accorsi, PhD2, John D. Idoine, PhD3, J. Anthony Parker, MD, PhD1, Jürgen T. Renze, PhD4, Richard C. Lanza, PhD5, John V. Frangioni, MD, PhD4 1
Division of Nuclear Medicine, Department of Radiology, Beth Israel Deaconess Medical Center, Boston, MA, 2Department of Radiology, The Children’s Hospital of Philadelphia, Philadelphia, PA, 3Department of Physics, Kenyon College, Gambier, OH, 4Division of Hematology/Oncology, Department of Medicine, Beth Israel Deaconess Medical Center, Boston, MA, 5Department of Nuclear Engineering, Massachusetts Institute of Technology, Cambridge, MA Purpose: Small animal radioscintigraphic imaging systems aim to achieve sub-millimeter resolution. At the present time, sub-millimeter calibration sources that can be placed at will within an imaged volume are not readily available. We have developed a method for producing technetium-99m (Tc-99m) sources in less than 15 minutes with readily available reagents. Procedures: Tc-99m pertechnetate [TcO4] was incubated with 45 m to 106 m diameter spherical anion exchange beads, washed, and mounted as desired for instrument calibration. Results: The procedure yields spherical sources having between 6.8 Ci to 11.1 Ci of Tc99m per source. This work shows that dual imaging of these sources using white light and radioscintigraphy permits measurement of system performance with high precision. Conclusion: Easily prepared, sub-millimeter Tc-99m spherical calibration sources are described, and it is demonstrated that such sources are useful for measuring the resolution and sensitivity of radioscintigraphic systems, such as those designed for small animal imaging. (Mol Imag Biol 2002;4:380–384) © 2002 Elsevier Science Inc. All rights reserved. Key Words: Radioscintigraphy; Single-photon emission computed tomography (SPECT); Technetium-99m; Calibration; Resolution; Coded aperture imaging.
Introduction
R
adioscintigraphy plays an important role in cancer detection and staging. A major focus of research in this area is the development of tumortargeted radiopharmaceuticals. This in turn requires preclinical testing in small animal models. Although clinical single photon emission computer tomography (SPECT) imaging systems typically have a maximum resolution of approximately 5–8 mm, newer small animal SPECT systems are reported to have resolutions approaching 830 m.1–3 Other high-resolution techniques, such as those employing pinholes, micro-collimators, and Compton cameras are also being developed. As a consequence, there is a general need for calibration sources, of the radioisotope under study that are far smaller than imaging system resolution and are of a known geometry. Address correspondence to: John V. Frangioni, M.D., Ph.D. HIM1023, Beth Israel Deaconess Medical Center 330 Brookline Avenue Boston, MA 02215. E-mail: [email protected]
380
Since there are several problems associated with SPECT imaging of small animals, including the need for relatively high doses of isotope, long scan times, and limited resolution, alternatives have been developed. One such technique utilizes a coded aperture, that is, a radio-opaque mask that is placed between the camera crystal and the object.4 The mask has a pattern of pinholes of fixed diameter, whose pattern is such that hundreds or even thousands of overlapping projections of the object fall on the detector crystal. Using straightforward algorithms based on the hole arrangement, the image is reconstructed. For many common objects, particularly those that are bright objects on a dark field, coded aperture imaging provides higher sensitivity, and depending on the design of the system, higher resolution and/or shorter scan times than SPECT.4 Like SPECT and other high-resolution techniques, there is a need for sub-millimeter calibration sources. In vivo radioscintigraphic imaging can be accomplished using targeted ligands associated with one of many possible radioisotopes. Some of the most com-
Sub-Millimeter Tc-99m Calibration Sources/ English et al. 381
mon isotopes used pre-clinically and clinically are iodine-123, indium-111, and Tc-99m. Of these, Tc-99m has many favorable characteristics including 140 keV monoenergetic gamma emission, a six hour half-life, low radiation absorbed dose, availability from a generator, and low cost. Indeed, nearly 80% of all radiopharmaceuticals used in clinical nuclear medicine are Tc99m-labeled compounds, and various strategies exist for labeling antibodies, peptides, and small molecules with it.5–9 In this work, a simple method for producing Tc-99m calibration sources having a diameter sufficiently small to permit characterization of imaging systems with submillimeter resolution is presented. In addition, the utility of these sources using a coded aperture imaging system is proven.
Materials and Methods Reagents AG1–X8 anion exchange resin in its formate form (catalog #140-1454) and Poly-Prep disposable plastic columns (catalog #731–1550) were purchased from BioRad (Hercules, CA). Tc-99m pertechnetate was eluted from a molybdenum-99 generator using normal saline. Water was purified to 18 M on a Milli-Q apparatus (Millipore, Bedford, MA).
periment, 10 L to 50 L of this slurry was placed on a Superfrost glass slide (Fisher Scientific, Hanover Park, IL). Individual beads were positioned using a razor blade and the slide was blotted with a paper towel to remove unwanted beads. A slide loupe or magnifying glass is used to ensure that the desired number of beads is on the slide and at the desired location. Excess water is then blotted and the slides are allowed to dry. For activity per bead/slide of greater than 10 Ci, measurement was made using a model CRC-15R dose calibrator (Capintec, Ramsey, NJ). For lower activities, or whenever system sensitivity was being measured, a Packard Cobra II well detector (Perkin Elmer Life Sciences, Boston, MA) was used.
White Light Imaging of Calibration Sources For single bead experiments, or those where bead separation was less than 5 mm, brightfield images were acquired using a calibrated Nikon (Melville, NY) Eclipse TE-300 microscope fitted with a Photometrics Sensys CCD Camera (Model 1401, Roper Scientific, Tucson, AZ). For experiments where bead separation was greater than 5 mm, a previously described small animal imaging system was used in darkfield mode.10 Image acquisition and analysis was performed using IPLab software (Scanalytics, Fairfax, VA).
Coded Aperture Radioscintigraphy Preparation of Calibration Sources All procedures were performed behind a 0.5-inch lead shield with a 0.25-inch leaded glass window (Biodex, Shirley, NY). Approximately 50 mCi of Tc-99m pertechnetate in 1 ml to 3 ml of normal saline was placed in a Poly-Prep column with the tip still attached. Poly-Prep columns have a porous 30 m polyethylene bed support that retains the anion exchange beads. One hundred milligrams of AG1-X8 resin was placed in a 1.5 ml Eppendorf tube, washed three times and resuspended in water at a concentration of approximately 200 beads per L. Five microliters (approximately 1000 beads) of this slurry was added to the Poly-Prep column, and the solution was pipetted up and down for two minutes to achieve constant bead collision with the Tc-99m pertechnetate. The tip of the Poly-Prep column was removed and the flow-through was discarded. Beads were washed with 20 ml of water, and the column was capped with the supplied yellow cap. One milliliter of water was added to the column, the beads were brought into suspension by up/down pipetting, and transferred to a 5 ml polystyrene test tube. The volume was adjusted to 4 ml with water and the beads were pelleted by centrifugation at 500 g for 30 seconds. The beads are barely visible on the bottom of the tube. All but approximately 200 L of the supernatant was removed, and beads were resuspended by up/down pipetting. Depending on the ex-
Coded aperture radioscintigraphy using a conventional Anger camera has been described in detail previously.4 Briefly, the mask was constructed from 1 mm thick sintered tungsten alloy, Catalog #K1750, manufactured by H.C. Starck Kulite Tungsten Products (East Rutherford, NJ). One point one mm holes were drilled in a 62 by 62 no-two-holes-touching pattern using a modified uniformly redundant array.4 The mask was fitted on a Forte gamma camera (ADAC Laboratories, Milpitas, CA) using a custom aluminum frame such that the detector to mask distance was 30 cm, the mask to object distance was 10 cm, the magnification factor was 3.0 and the field of view was 9.3 cm. Five hundred thousand total counts were acquired in a 512 by 512 matrix. Reconstructions were performed using MatLab v. 6.1.0.450, release 12.1 (MathWorks, Natick, MA) running on a 550 MHz Pentium III processor.
Results Calibration Source Preparation and Uniformity AG1–X8 anion exchange resin is composed of spherical beads with wet diameters in the range 45 m to 106 m. Using the protocol described in Materials and Methods and shown in Figure 1, individual beads were loaded with Tc-99m pertechnetate. In four separate ex-
382 Molecular Imaging and Biology, Volume 4, Number 5
the length of time, and/or the amount of agitation of the initial binding reaction. The relatively low percent bound was acceptable to us, however, since one of the goals was to prepare calibration sources in as short a time as possible. The entire procedure took less than 15 minutes. It should be noted that the capacity of the beads (1.2 mEq per ml) was more than 1500 times higher than the number of moles actually loaded.
Single-Source Measurements
Figure 1. Calibration Source Preparation. The flowchart shows the steps taken, as described in detail in Materials and Methods, for preparing sub-millimeter calibration sources.
periments, the amount of activity per bead was between 6.8 Ci to 11.1 Ci. For a typical labeling of 1000 beads with 50 mCi of Tc-99m pertechnetate using the protocol described in Materials and Methods, 13.6% to 22.6% of total activity was bound to the beads. This number was easily manipulated by changing the number of beads,
Since the diameter of individual beads varied widely, beads to be used for size standards were adsorbed to glass slides and their diameter measured precisely using light microscopy (see Materials and Methods). After air drying, the strength of bead adsorption was such that beads did not move during transport, shaking, or light blowing. Figure 2 shows the results from a typical experiment. In this case, the single calibration bead had a measured diameter of 96.3 m. Coded aperture reconstruction of the image took approximately 812 msec and is also displayed in Figure 2. By projecting a line through the center of the bead, the system resolution, that is, the full-width half-maximum (FWHM) of intensity, could be measured. The theoretical system resolution, which is a function of the mask and camera characteristics, was 1.67 mm. The FWHM measured after coded aperture reconstruction of the single 96.3 m source was 1.70 0.37 mm (Figure 2). It is interesting to note that the theoretical system resolution of 1.67 mm was obtained from an advanced mathematical model of the point spread function of
Figure 2. Single-Source Instrument Calibration. A single anion exchange bead measuring 96.3 m was labeled with 11.1 Ci of Tc-99m and imaged using brightfield microscopy (left panel). The same bead was imaged with a coded aperture imaging system, and the tomographic reconstruction through the center of the bead was displayed (middle panel; bead at arrow). Intensity (arbitrary units) is shown as an insert. The horizontal line projection through the bead’s center (right panel) was used to determine system resolution (FWHM 1.70 0.37 mm).
Sub-Millimeter Tc-99m Calibration Sources/ English et al. 383
the system.11 Use of the classic formula, where the geometric (1.1 mm * 4 / 3 1.47 mm) and system (3.7 mm / 3 1.23 mm) resolution add in quadrature, gives a system resolution of 1.92 mm. Using a small point source was important for an experimental validation of the prediction of the more advanced model with no need to deconvolve the size and shape of the point source that are often not known with sufficient accuracy. Single source measurements are also quite useful for measuring system sensitivity. Prior to imaging, a glass slide with a single Tc-99m bead was counted as described in Materials and Methods. The same bead was then imaged as described for Figure 2 and its count rate recorded. One of the main advantages of coded aperture imaging is its increased sensitivity over methods such as SPECT. Indeed, a single bead loaded with 11.1 Ci of Tc-99m yielded count rates on our system of approximately 2200 cps (11,892 cpm/Ci), with the 500,000 total counts desired for image reconstruction being acquired in less than four minutes.
Multiple-Source Measurements Single Tc-99m labeled sources can be positioned anywhere in three dimensions. Shown in Figure 3 is a planar arrangement of three beads in an asymmetrical
triangle. Beads are labeled A, B, and C, with the distances between beads A-B, A-C, and B-C being 9.3 mm, 12.7 mm, and 11.3 mm, respectively, as measured by darkfield illumination. The distances measured using coded aperture imaging and reconstruction were 9.3 mm, 13.0 mm, and 11.2 mm. These are within the error associated with the size of an individual pixel ( 0.37 mm).
Discussion The goal of this work was to produce sub-millimeter Tc-99m sources for measuring the resolution and sensitivity of small animal imaging systems. The procedure requires less than 15 minutes to complete and uses readily available, relatively inexpensive reagents. Although the sources described in this study utilized Tc-99m pertechnetate, the procedure should be readily adaptable to other anionic radioisotopes. To be used with cationic radioisotopes, one need only to use a cation exchange resin in place of the anion exchange resin. The geometry of the beads used in this study is of particular importance. Despite being dried on a glass slide, the beads remain essentially spherical as evidenced by both light microscopy and coded aperture imaging in orthogonal planes (data not shown). For ultra-high resolution systems, spherical source geometry
Figure 3. Multiple-Source Instrument Calibration. Three anion exchange beads (arrows), each measuring between 88 m to 98 m, were loaded with approximately 11.1 Ci of Tc-99m per bead and imaged using darkfield illumination (left panel). A black circle was drawn around each bead on the slide with a marker. The coded aperture reconstruction of the three beads is shown on the right panel. Beads are labeled A, B, and C, as described in the text. Intensity (arbitrary units) is shown as an insert.
384 Molecular Imaging and Biology, Volume 4, Number 5
should permit detection of system inhomogeneities that might only be seen with a particular detector position. In this study, multiple beads were placed in a single plane (glass slide). The bead format, however, permits custom holders to be constructed that position the sources anywhere in three-dimensional space, thus permitting testing of complex geometries. Finally, although we imaged the calibration sources using a coded aperture system, they should perform similarly well in SPECT, and other types of radioscintigraphic imaging systems.
3.
4.
5. 6.
Conclusion We have developed a simple method for preparing submillimeter, spherical Tc-99m sources. Such sources may be quite useful for calibrating the next generation of high-resolution, high-sensitivity radioscintigraphic small animal imaging systems. We thank Alec M. DeGrand for technical assistance and Grisel Rivera for administrative assistance. RA and RCL were funded by the Office of National Drug Control Policy Contract #DAAD07-98-C-0117 and the Federal Aviation Administration Grant #93G-053. JTR was supported by a fellowship within the Postdoc-Programme of the German Academic Exchange Service (DAAD). JI and JVF were funded by grant #R21CA88870 from the National Institutes of Health.
7.
8. 9.
10.
References 1. Habraken, J.B.; de Bruin, K.; Shehata, M.; et al. Evaluation of high-resolution pinhole SPECT using a small rotating animal. J. Nucl. Med. 42:1863–1869; 2001. 2. Liu, Z.; Kastis, G.A.; Stevenson, G.D.; et al. Quantitative analysis of acute myocardial infarct in rat hearts with is-
11.
chemia-reperfusion using a high-resolution stationary SPECT system. J. Nucl. Med. 43:933–999; 2002. Acton, P.D.; Kung, M.P.; Hou, C.; Plossl, K.; Keeney, C.; Kung, H.F. Ultra-high resolution single photon emission tomography imaging of the mouse striatum. Eur. J. Nucl. Med. Mol. Imaging. 29:446; 2002. Accorsi, R.; Gasparini, F.; Lanza, R.C. A coded aperture for high-resolution nuclear medicine planar imaging with a conventional anger camera: experimental results. IEEE Trans. Nucl. Sci. 48:2411–2417; 2001. Dilworth, J.R.; Parrott, S.J. The biomedical chemistry of technetium and rhenium. Chem. Soc. Rev. 27:43–55; 1998. Mahmood, A.; Kuchma, M.H.; Freiberg, E.; Goldstone, J.; Davison, A.; Jones, A.G. Functionalized tetradentate N2S2 chelates and their technetium-99m and rhenium complexes: synthesis, spectroscopy and structural characterization. In: Nicolini, M., Mazzi, U., eds. Technetium, rhenium and other metals in chemistry and nuclear medicine. Padova, Italy: Servizi Grafici Editoriali; 1999:253–257. Mahmood, A.; Kuchma, M.H.; Goldstone, J.; Morse, C.; Davison, A.; Jones, A.G. An N2S2 tetradentate chelate for solid-phase synthesis: evaluation in solution and solid phase and characterization of technetium-99 complexes. In: Nicolini, M., Mazzi, U., eds. Technetium, rhenium and other metals in chemistry and nuclear medicine. Padova, Italy: Servizi Grafici Editoriali; 1999:71–76. Jurisson, S.S.; Lydon, J.D. Potential technetium small molecule radiopharmaceuticals. Chem. Rev. 99:2205–2218; 1999. Liu, S.; Edwards, D.S. 99mTc-labeled small peptides as diagnostic radiopharmaceuticals. Chem .Rev. 99:2235–2268; 1999. Zaheer, A.; Lenkinski, R.E.; Mahmood, A.; Jones, A.G.; Cantley, L.C.; Frangioni, J.V. In vivo near-infrared fluorescence imaging of osteoblastic activity. Nat Biotechnol. 19:1148–1154; 2001. Accorsi, R., Design of near-field coded aperture cameras for high-resolution medical and industrial gamma ray imaging. Ph.D. Thesis, Nuclear Engineering, Massachusetts Institute of Technology: Cambridge, MA. 2001.