Performance improvement of small gamma camera using NaI(Tl) plate and position sensitive

Transcription

1 Home Search Collections Journals About Contact us My IOPscience Performance improvement of small gamma camera using NaI(Tl) plate and position sensitive photo-multiplier tubes This article has been downloaded from IOPscience. Please scroll down to see the full text article Phys. Med. Biol ( View the table of contents for this issue, or go to the journal homepage for more Download details: IP Address: The article was downloaded on 05/02/2011 at 04:34 Please note that terms and conditions apply.

2 INSTITUTE OF PHYSICS PUBLISHING Phys. Med. Biol. 49 (2004) PHYSICS IN MEDICINE AND BIOLOGY PII: S (04) Performance improvement of small gamma camera using NaI(Tl) plate and position sensitive photo-multiplier tubes Myung Hwan Jeong, Yong Choi, Yong Hyun Chung, Tae Yong Song, Jin Ho Jung, Key Jo Hong, Byung Jun Min, Yearn Seong Choe, Kyung-Han Lee and Byung-Tae Kim Department of Nuclear Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, 50 Ilwon-Dong, Kangnam-Ku, Seoul, , Korea Received 20 July 2004, in final form 20 September 2004 Published 15 October 2004 Online at stacks.iop.org/pmb/49/4961 doi: / /49/21/008 Abstract The purpose of this study was to improve the performance of a small gamma camera, utilizing a NaI(Tl) plate and a 5 position sensitive PMT. We attempted to build a NaI(Tl) plate crystal system which retained all its advantages, while at the same time integrating some of the advantages inherent in an array-type scintillation crystal system. Flood images were obtained with a lead hole mask, and position mapping was performed by detecting hole positions in the flood image. Energy calibration was performed using the energy spectra obtained from each hole position. Flood correction was performed using a uniformity correction table containing the relative efficiency of each image element. The spatial resolution was improved about 16% after correction at the centre field of view. Resolution deterioration at the outer field of view (OFOV) was considerably ameliorated, from 6.7 mm to 3.2 mm after correction. The sensitivity at the OFOV was also increased after correction, from 0.7 cps µci 1 to 2.0 cps µci 1. The correction also improved uniformity, from 5.2% to 2.1%, and linearity, from 0.5 mm to 0 mm. The results of this study indicate that the revised correction method can be employed to considerably improve the performance of a small gamma camera using a NaI(Tl) plate-type crystal. This method also provides high spatial resolution and linearity, like array-type crystals do, while retaining the specific advantages of plate-type crystals. 1. Introduction Recently, interest has grown regarding compact and high resolution small gamma cameras. These devices can be used for applications ranging from patient studies, such as scintimammography, to small animal imaging. Several groups have been working on the /04/ $ IOP Publishing Ltd Printed in the UK 4961

3 4962 M H Jeong et al development of the small gamma camera using position sensitive photo-multiplier tubes (PSPMT), and scintillation crystals (Pani et al 1998, Kim et al 2000, Williams et al 2000). Images of these small gamma cameras should have high uniformity and linearity. However, if a fixed pulse-height window is used over the entire field of view (FOV) of a camera, different energy responses from different points in the field of view can result in different shapes in energy spectra. This, in turn, can cause apparent discrepancies in sensitivity in different parts of the detector, essentially leading to image non-uniformities (Steinbach et al 1997). Furthermore, PSPMTs inherently provide nonlinear and non-uniform responses (Kume et al 1986). For these reasons, it has been difficult to achieve appropriate linearity and uniformity in small gamma camera systems which utilize scintillation crystals and PSPMTs. In order to obviate these problems, several correction methods have been developed, for example, the multi-wire read-out technique, in which the charge on each anode wire is individually read out and digitized (He et al 1993, Bird et al 1994, Truman et al 1994), or maximum-likelihood position estimation (MLPE), which involves the use of a look-up table and the maximum-likelihood method to map from measured PMT signals to position estimates (Gray and Macovski 1976, Joung et al 2001, Joung et al 2000, Chung et al 2004). All these methods were demonstrated to improve linearity and spatial resolution. Correction methods, utilizing array crystal mapping and the relative efficiency of each image pixel, were also developed in order to improve performance (Steinbach et al 1997, Wojcik et al 1998, Schramm et al 2000, Wojcik et al 2001). Among these correction methods, the approach of position mapping, energy calibration and flood correction procedures developed by Wojcik et al. (Steinbach et al 1997, Wojcik et al 1998, Wojcik et al 2001) represents one of the more practical and efficient methods due to its ease of implementation regarding hardware, and speed in computation time. However, this correction method has been applied only to an array-type crystal, and has not yet been used for a plate-type crystal since each pixel region requires separation in order to map the continuous output of the PSPMT onto detector pixels. In order to obtain high resolution images, an array-type scintillation crystal is usually used. When a gamma ray is absorbed by one of the tiny crystals of an array-type scintillator, almost all the induced scintillation light is collected in a small area of a PSPMT. Therefore, the tinier the crystal used in an array-type scintillation crystal system is, the higher the obtained spatial resolution will be. However, an array-type scintillation crystal has disadvantages, such as lower sensitivity, lower energy resolution and higher cost than a plate-type scintillation crystal caused by the gaps between the crystal elements and small pixel size. The purpose of this study was to improve the performance of a small gamma camera utilizing a NaI(Tl) plate and a 5 position sensitive PMT. We attempted to build a NaI(Tl) plate crystal system which retained all its advantages, while at the same time integrating some of the advantages inherent in an array-type scintillation crystal system. We hypothesized that the revised approach would provide comparable spatial resolution and linearity with an arraytype system, while retaining the advantages of the gamma camera with a plate-type crystal system. The effectiveness of the revised correction method was evaluated by comparing the performance of a NaI(Tl) plate system using the modified correction method, with the performance of a CsI(Tl) array system using the correction method developed by Wojcik et al (Wojcik et al 1998, Schramm et al 2000, Wojcik et al 2001). 2. Materials and methods 2.1. System configuration The small gamma camera consisted of a general-purpose parallel hole collimator, a scintillation crystal, a 5 PSPMT (Hamamatsu R3292) and subtractive resistive read-out electronics

4 Performance improvement of small gamma camera 4963 Figure 1. The basic components of a small gamma camera system with a lead hole mask used for position mapping. (figure 1, Kim et al 2000, Wojcik et al 2001). The scintillation crystal employed in this study was a NaI(Tl) plate, 120 mm in diameter and 6 mm in thickness. The parallel-hole collimator was 24 mm in length, with a hole diameter of 1.5 mm and a septum thickness of 0.2 mm. The 28-X by 28-Y outputs were coupled to preamplifiers and a resistive charge divider, a set-up which is described in detail in Clancy et al (1997). Fifty-six signals from the R3292 PSPMT were reduced to four signals. The four signals were amplified and digitized using ADC with 40 Msamples s 1 sampling rate then used to localize an event employing the Anger logic. The data acquisition programs were based on Kmax (Sparrow Corporation, Port Orange, FL) Correction methods Position mapping. Our position mapping procedure was developed in order to correct the distorted hole positions in the image obtained by a NaI(Tl) plate scintillation crystal. A lead hole mask (150 mm 150 mm 4mm,1mmholediameter,5mmpitch)was placed in contact with the NaI(Tl) plate crystal without a collimator (figure 1). To achieve 2.5 mm hole-to-hole distance resolution, raw images were acquired four times at the locations (0, 0), (2.5 mm, 0), (0, 2.5 mm) and (2.5 mm, 2.5 mm). Raw images were obtained applying the Anger logic for 1 h at each position generating about counts/pixel, with a high applied voltage of 950 V, using a 99m Tc point source (500 µci) located at a distance of at least five times the largest dimension of the useful field of view (UFOV) above the detector. The obtained raw images were overlapped and the position mapping table was generated by using the overlapped raw images. Figure 2 shows the raw lead hole mask image and the generated map (40 40) of hole positions of a NaI(Tl) plate system Energy calibration. Energy calibration was performed using the pulse height spectra obtained from each hole position. An energy spectrum table, containing 1600 pulse height spectra, was obtained for 10 h (about 800 count at a peak of a pulse spectrum) with a generalpurpose parallel hole collimator using a point source (2 mci 99m Tc) located 40 cm above the detector. An energy window was applied to each individual energy spectrum obtained from each hole position. Figure 3 shows the energy spectra obtained from three different positions of the NaI(Tl) plate system. The energy spectrum was not only truncated, but also shifted as the measured location moved to off-centre of the FOV Flood correction. Flood correction was performed following the previously reported procedure (Steinbach et al 1997) using a uniformity correction table, listing the relative

5 4964 M H Jeong et al (a) (b) Figure 2. Raw image (a) and position mapped image (b) obtained with NaI(Tl) plate system. Figure 3. Energy spectra of holes mapped by lead hole mask located in different regions in NaI(Tl) plate system. These were recorded with a 99m Tc point source that was kept in a fixed location. Spectra were taken from the centre (top) and off-centre regions (middle and bottom) of the detector.

6 Performance improvement of small gamma camera 4965 (a) (b) Figure 4. Uniformity correction table image (a), and uniformity corrected image (b). efficiency of each imaging element. The uniformity correction table was obtained for 10 h with a flood source and contained large counts about count/pixel to minimize the statistical noise. The flood source was sufficiently large to cover the entire detector, and was placed in direct contact with the collimated detector. The flood source phantom consisted of a plastic container filled with radioactive solution (2 mci 99m Tc). Since the counts of each pixel in an obtained image were much fewer than the corresponding values of the uniformity correction table, the image was multiplied by a scaling factor (1000) and then divided by the obtained uniformity correction table. Figures 4(a) and (b) demonstrate a flood correction table image and the corrected image Performance measurements Spatial resolution. Spatial resolution was measured using two capillary tubes having 50 µci 99m Tc each with inner diameter measurements of 0.4 mm. Two-line sources with 8 mm spacing were used to obtain a millimetres-per-pixel calibration. The line source images were obtained as a function of the distance from detector to source, which vertically varied from 0 mm to 60 mm and as a function of the distance from the centre to the edge of the detector, which varied horizontally from 0 mm to 30 mm at the detector surface. The images were acquired with a 20% energy window. Full width at half-maximums (FWHMs) of the profiles of the two-line images were measured Sensitivity. System sensitivity was measured using a 50 µci 99m Tc point source image, obtained at a range of distances from the detector to source, varying from 0 mm to 40 mm, and a range of distances from the centre to the edge of the detector, which varied from 0 mm to 30 mm. Radioactivity decay and background activities were corrected when the sensitivity was measured. Uniformity correction was not applied and the images were acquired with a 20% energy window Linearity. Linearity was measured with a parallel-line bar phantom, and was expressed as a standard deviation of the line spread function peak separation. After the image of the parallel-line bar phantom was obtained, the linearity was calculated using the line image.

7 4966 M H Jeong et al Distance from detector to source (cm) (a) Distance from centre to source (cm) (b) Figure 5. System resolutions measured at varying distance from detector to source (a) and from centre to source (b) Uniformity. The uniformity was measured using a uniformity phantom that was filled with radioactive solution (500 µci 99m Tc). The uniformity of the NaI(Tl) plate systems was calculated after smoothing with a 3 3 Gaussian filter. Integral uniformity (a maximum deviation) and differential uniformity (a maximum rate of change over a specified distance) were also measured (NEMA 1994) Comparison of performances In order to assess the effectiveness of the method revised in this study using NaI(Tl) plate, the performance of the small gamma camera employing the CsI(Tl) array (120 mm diameter, pixel size 2 mm 2mm 3 mm) and the correction method developed by Wojcik et al was measured. The performance characteristics of the CsI(Tl) array system, such as resolution, sensitivity, linearity and uniformity, were measured after correction, and compared with the performance characteristics of the NaI(Tl) plate system applying the modified correction method. 3. Results 3.1. Spatial resolution The resolution of the NaI(Tl) plate system was improved by about 16% using the correction method, and was similar to that of the CsI(Tl) array system after correction (figure 5(a)). Figure 5(b) demonstrates that the resolution deterioration observed in the NaI(Tl) plate system at a distance of 30 mm off-centre was considerably improved, from 6.7 mm FWHM to 3.2 mm FWHM, after correction. The resolution of the NaI(Tl) plate system after correction was shown to be nearly constant, very much like that of the CsI(Tl) array system when the distance between the source and the centre of the detector increased Sensitivity The sensitivity of the NaI(Tl) plate system at the centre of the FOV remained similar both before and after correction (figure 6(a)). However, the sensitivity of the NaI(Tl) plate system at 30 mm off-centre considerably increased after correction, from 0.7 cps µci 1 to 2.0 cps µci 1 (figure 6(b)). The sensitivity of the NaI(Tl) plate system was considerably better than that of

8 Performance improvement of small gamma camera 4967 System sensitivity (cps/uci) System sensitivity (cps/uci) Distance from detector to source (cm) (a) Distance from centre to source (cm) (b) Figure 6. System sensitivities measured at varying distance from detector to source (a) and from centre to source (b). Figure 7. Line source images without (left) and with (right) the corrections obtained with the NaI(Tl) plate system. Table 1. Linearities of NaI(Tl) plate and CsI(Tl) array systems. Linearity NaI(Tl) plate (mm) CsI(Tl) array (mm) Before correction Centre mm off-centre After correction Centre mm off-centre 0 0 the CsI(Tl) array system over the entire FOV after correction (NaI(Tl) plate: 3.4 cps µci 1, CsI(Tl) array: 1.4 cps µci 1 at the centre) Linearity The linearity of the NaI(Tl) plate system improved after correction, from 0.5 mm to 0 mm at the centre of the FOV, and from 1.5 mm to 0 mm, at 35 mm off-centre. Before correction, the linearity of the NaI(Tl) plate system was slightly worse than that of the CsI(Tl) array system, but after correction the linearity of both systems was about the same (table 1). Figure 7 shows images ( ) of a parallel-line bar phantom obtained with the NaI(Tl) plate system.

9 4968 M H Jeong et al Figure 8. Flood source images without (left) and with (right) the corrections obtained with the NaI(Tl) plate system. Table 2. Uniformities of NaI(Tl) plate and CsI(Tl) array systems. Uniformity NaI(Tl) plate (%) CsI(Tl) array (%) Before correction Integral Differential After correction Integral Differential Uniformity The integral and differential uniformity of the NaI(Tl) plate system both improved after correction, from 9.7% to 5.2% and from 3.6% to 2.1%, respectively. The uniformity of the NaI(Tl) plate system was better than that of the CsI(Tl) array system after correction (table 2). Figure 8 shows images of a flood source ( ) for the NaI(Tl) plate system, both before and after the correction. 4. Discussion and conclusion We developed a method which maps the detected gamma ray positions in a NaI(Tl) plate crystal. In order to evaluate the performance characteristics conferred by the revised method employing the NaI(Tl) plate, resolution, sensitivity, linearity and uniformity were assessed both before and after correction, and were compared to those of the array-type scintillator system using the correction methods developed by Wojcik et al. The resolution of the NaI(Tl) plate system at the outer part of the FOV was considerably improved, by up to 52% after correction. The energy spectra, defined by the hole positions of the lead hole mask in the NaI(Tl) plate system, were different from each other in the photopeak positions (figure 3). Thus the photopeak of an event might not exactly enter into the fixed energy window defined by the spectrum obtained at the centre of the FOV, when an event occurred at the edge of the detector. This problem of deteriorating resolution and linearity was corrected by using the energy calibration method, and by applying the appropriate energy window at each imaging element. In this manner, the performance characteristics of the NaI(Tl) plate system were improved, as illustrated in figures 5, 6, 7 and 8.

10 Performance improvement of small gamma camera 4969 The sensitivity of the NaI(Tl) plate system at the edge of the detector was also improved, by a factor of almost 3 after correction. Figure 6 shows that the sensitivity of the NaI(Tl) plate system was considerably higher than that of the CsI(Tl) array system. The degradation of sensitivity was still observed, however, at the edge of the NaI(Tl) system, even after correction, because the hole positions overlapped, and were not identifiable at the far edge (figure 6(b)). Figure 8 shows that the uniformity of the NaI(Tl) plate system was considerably improved after correction. This result means that the revised NaI(Tl) plate system can minimize the risk of false positive and false negative results with hot lesions. While the resolution and linearity of the NaI(Tl) plate system were similar to those of the CsI(Tl) array system, sensitivity and uniformity were better in the NaI(Tl) plate system. The results of this study indicate that the modified correction method considerably improves the performance of a small gamma camera using NaI(Tl) plate-type crystal and can help achieve high spatial resolution and linearity, like array-type crystals while retaining the advantages of plate-type crystals. Acknowledgments This study was supported by Korea Institute of Science & Technology Evaluation and Planning (KISTEP) and Ministry of Science & Technology through their National Nuclear Technology Program and by a grant of the Korea Health 21 R&D Project (02-PJ3-PG6-EV ), Ministry of Health & Welfare, Republic of Korea. References Bird A J, He Z and Ramsden D 1994 Multi-channel readout of crossed-wire anode photomultipliers Nucl. Instrum. Methods A Chung Y H, Choi Y, Song T Y, Jung J H, Cho G, Choe Y S, Lee K-H, Kim S E and Kim B T 2004 Evaluation of maximum-likelihood position estimation with Poisson and Gaussian noise models in a small gamma camera IEEE Trans. Nucl. Sci Clancy R L, Thompson C J, Robar J L and Bergman A M 1997 A simple technique to increase the linearity and field-of-view in position sensitive photomultiplier tubes IEEE Trans. Nucl. Sci Gray R M and Macovski A 1976 Maximum a posteriori estimation of position in scintillation cameras IEEE Trans. Nucl. Sci He Z, Bird A J and Ramsden D 1993 A 5 inch diameter position-sensitive scintillation count IEEE Trans. Nucl. Sci Joung J H, Miyaoka R S, Kohlmyer S and Lewellen T K 2000 Implementation of ML based positioning algorithms for scintillation cameras IEEE Trans. Nucl. Sci Joung J H, Miyaoka R S and Lewellen T K 2001 cmice: a high resolution animal PET using continuous LSO with a statistics based positioning scheme IEEE Medi. Imag. Conf. Conf. Rec. pp M3 5 Kim J H, Choi Y, Joo K-S, Sihn B S, Chong J W, Kim S E, Lee K H, Choe Y S and Kim B T 2000 Development of a miniature scintillation camera using NaI(Tl) scintillator and PSPMT for scintimammography Phys. Med. Biol Kume H, Muramatsu S and Lida M 1986 Position sensitivity photomultiplier tubes for scintillation imaging IEEE Trans. Nucl. Sci National Electrical Manufacturers Association 1994 Performance measurements of scintillation cameras Standards Publication No. NU (Washington, DC: NEMA). Pani R, De Vincentis G, Scopinaro F, Pellegrini R, Soluri A, Weinberg I N, Pergola A, Scafe R and Trotta G 1998 Dedicated gamma camera for single photon emission mammography (SPEM) IEEE Trans. Nucl. Sci Schramm N, Wirrwar A, Sonnenberg F and Halling H 2000 Compact high resolution detector for small animal SPECT IEEE Trans. Nucl. Sci Sparrow Corporation, Daytona Beach

11 4970 M H Jeong et al Steinbach D, Majewski S, Williams M, Kross B, Weisenberger A G and Wojcik R 1997 Development of a small field of view scintimammography camera based on a YAP crystal array and a position sensitive PMT IEEE Med. Imag. Conf. Conf. Rec. pp Truman A, Bird A J, Ramsden D and He Z 1994 Pixellated CsI(Tl) arrays with position-sensitive PMT readout Nucl. Instrum. Methods A Williams M B, Goode A R, Galbis-Reig V, Majewski S, Weisenberger A G and Wojcik R 2000 Performance of a PSPMT based detector for scintimammography Phys. Med. Biol Wojcik R, Majewski S, Kross B, Popov V and Weisenberger A G 2001 Optimized readout of small gamma cameras for high resolution single gamma and positron emission imaging IEEE Medi. Imag. Conf. Conf. Rec. pp M10 1 Wojcik R, Majewski S, Kross B, Steinbach D and Weisenberger A G 1998 High spatial resolution gamma imaging detector based on a 5 diameter R3292 Hamamatsu PSPMT IEEE Trans. Nucl. Sci