Development for a novel phantom for evaluating image quality in small-animal single photon emission computed tomography and positron emission tomography.

doi:10.21203/rs.3.rs-4916789/v1

Download PDF

Research Article

Development for a novel phantom for evaluating image quality in small-animal single photon emission computed tomography and positron emission tomography.

https://doi.org/10.21203/rs.3.rs-4916789/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Objectives

The National Electrical Manufacturers Association (NEMA) has released guidelines delineating the performance of positron emission tomography (PET) devices designed for small animals. However, the NEMA NU 4 image quality phantom could not measure the known contrasts of the hot rod images and the recovery coefficient (RC) of cold rod images due to the structure of the phantom. Thus, we have devised novel hot rod and cold rod phantoms capable of evaluating uniformity and RCs for both hot rod and cold rod images. This study aimed to assess uniformity, image contrasts, and RCs in hot rod and cold rod images of single photon emission computed tomography (SPECT) and PET using the newly developed phantom.

Methods

The new physical phantom consisted of rod and pool sections. To assess image uniformity, the pool section, designed in a cylindrical shape, was utilized. Conversely, the rod section was created in hot rod and cold rod shapes and integrated into a cylindrical phantom with the same design as the pool section. Hot rod and cold rod phantoms were designed with six different 1–6 mm diameter rods. The rod and pool sections of the hot rod phantom were separately filled with ^99mTc and ¹⁸F solutions. In the rod section, the cylindrical part was defined as the background (BG), with a radioactive concentration ratio of 4:1 for the hot rod and BG. The cylindrical part containing the cold rod was separately filled with ^99mTc and ¹⁸F solutions. The ^99mTc and ¹⁸F phantoms were acquired separately over 30 min. A transverse image with a cubic voxels of 0.8 mm length was reconstructed using a pixel-based ordered subset expectation maximization algorithm.

Results

The contrast of the hot rod for ^99mTc and ¹⁸F showed lower values with a decreasing rod diameter. Furthermore, the ^99mTc image demonstrated a higher contrast than the ¹⁸F image and approached the true contrast. The cold rod contrasts with ^99mTc and ¹⁸F followed a similar trend as the hot rod contrast. The RCs for the hot rods with 4–6 mm diameters were similar, whereas hot rods with diameters ≤ 3 mm revealed lower values as the rod diameter decreased. The inverse RC was lower with a decreasing cold rod diameter. Moreover, the cold rod image with ¹⁸F demonstrated a lower inverse RC than with the ^99mTc. The percent coefficient of variation (%CV) for the ^99mTc and ¹⁸F images was 4% and 7%, respectively, with the ^99mTc image displaying a lower %CV compared to the ¹⁸F image.

Conclusion

We have developed a new phantom that allows physical phenomenon evaluation in small animal SPECT and PET images, and can evaluate the image contrast, RC, and uniformity of both hot rod and cold rod images.

partial volume effect

small-animal

SPECT

PET

cold rod phantom

Small animal imaging in nuclear medicine plays a crucial role in drug development and disease elucidation [1, 2]. The quantitative assessment of nuclear medicine images using single-photon emission computed tomography (SPECT) and positron emission tomography (PET) is pivotal for assessing drug efficacy and understanding pathological conditions at the molecular level. SPECT and PET scanners may demonstrate performance variations owing to factors such as scintillators, collimators, acquisition, and image reconstruction parameters, and over time, this variability can impact quantitative values, potentially resulting in image artifacts [3]. The National Electrical Manufacturers Association (NEMA) has published guidelines for evaluating the performance of PET devices for small animals [4]. The NEMA NU 4 image quality phantom (Fig. 1) is employed to assess the image quality of SPECT and PET images, allowing physical indicator evaluation such as uniformity, recovery coefficients (RCs) of hot rods, contrast, and accuracy correction [5–8]. However, the NEMA NU 4 image quality phantom failed to accurately measure the known contrasts of hot rod images due to the inability to fill the radioactive solution around the hot rod section. To interpret the quantitative values, accurate contrast measurements are essential. In addition, the NEMA NU 4 phantom could assess the RCs of hot rod images but could not evaluate the RCs of cold rod image. The evaluation of cold rod images is equally important as the evaluation of hot rod images, as highlighted in several previous studies [9–13]. Cold rod image evaluation must be understood to interpret the pathophysiology and quantitative metrics in nuclear cardiology and neurology.

A new phantom capable of assessing uniformity and RCs of both hot rod and cold rod images was developed, addressing the limitations of the NEMA NU 4 phantom. In the same outer container, the new phantom can be used as a hot rod phantom or cold rod phantom. Furthermore, the hot rod phantom was used to assess known contrasts by being filled with two different radioactive solutions. To evaluate the image quality and the accuracy of quantitative metrics, the availability of the known contrasts is vital. This study aimed to assess uniformity, image contrasts, and RCs for hot rod and cold rod images with ^99mTc and ¹⁸F using this novel phantom.

Phantom design

The new phantom manufactured by Hokuriku EP Co., Ltd. was comprised of rod and pool sections (Fig. 2). The pool section, designed in a cylindrical shape (40 mm diameter and 20 mm length), was utilized to evaluate image uniformity. The rod section was created in hot rod and cold rod shapes and integrated into a cylindrical phantom with a same design as the pool section. In the rod section, hot rod and cold rod parts were designed with six different rod sizes ranging from 1 to 6 mm in diameter (length: 20 mm). Acrylic tubes and acrylic rods without cavities, respectively, were used to construct the hot rod and cold rod parts, and hot rod and cold rod parts can be exchanged in the same pool section. To reduce the impact on image quality from radiation interference, such as contamination of photons or scattering from another section, a 27 mm acrylic cap is attached between the pool and rod sections. Since the full width at half maximum (FWHM) values of an index of the spatial resolution of commercial small-animal SPECT or PET scanners were approximately 1–2 mm for both ¹⁸F and ^99mTc, the acrylic cap thickness is designed to be at least 10 times the FWHM value [15–18]. This design also prevents mutual interference between the images of the rod and pool sections owing to the partial volume effects (PVE). If radiation interference affects the image and quantitative assessments, the rod and pool sections can be effectively separated.

Phantom preparation

The hot rod phantoms with ^99mTc and ¹⁸F were separately created. Hot rods were assembled into the cylindrical holder that was utilized as the background (BG). The radioactive concentrations of the hot rod and the BG part in the rod section were set at 8 and 2 MBq/mL with ^99mTc, or 48 and 12 MBq/mL with ¹⁸F, respectively. The radioactive concentration ratio of the hot rod to BG was set at 4:1. Moreover, the pool section of the hot rod phantom was set at 8 MBq/mL with ^99mTc or 48 MBq/mL with ¹⁸F, respectively.

The cold rod phantom was inserted into the rod section, and the rod section was filled with a ^99mTc solution of 8 MBq/mL or a ¹⁸F solution of 48 MBq/mL, respectively. Additionally, the pool section was filled with a nonradioactive water as BG. Hot rod and cold rod phantoms were separately prepared.

Acquisition and image reconstruction parameters

The SPECT-PET/computed tomography (CT) scanner was utilized as a VECTor⁺/CT (MILabs B.V., Netherlands) of a triple-detectors gamma camera with high-energy ultrahigh-resolution rat and mouse-type clustered multi-pinhole collimator. The ^99mTc and ¹⁸F phantoms were acquired separately over 30 min. The photon energy window for ^99mTc was set at 141 keV ± 10%, with a 7% scatter window on each side. A transverse image with a cubic voxels of 0.8 mm length was reconstructed using a pixel-based ordered subset expectation maximization algorithm. [19]. The numbers of subset and iteration were set to 32 and 8 for ^99mTc, and 32 and 17 for ¹⁸F, respectively. A Gaussian filter, with a FWHM of 1.6 mm, was applied as a postfilter for both ^99mTc and ¹⁸F. Attenuation and scatter corrections were conducted using the triple energy window technique and CT-based attenuation correction [20].

Image assessment

Image analysis was performed using PMOD software (version 4.002, PMOD Technologies LLC, Fallanden, Switzerland). Circular regions of interest (ROIs) were drawn on hot rods and cold rods with diameters ranging from 1 to 6 mm. Furthermore, a circular ROI with a 12-mm diameter was also drawn in the BG region inside the 1–6 mm rod parts (Fig. 3). The hot rod and cold rod image contrasts were calculated by the following equations (1) and (2):

Contrast_hot= \(\:\frac{B}{A}\) (1)

Contrast_cold= \(\:\frac{A-B}{A}\) × 100 (2)

where A and B represent the average values of each rod and the average value of the BG part. RCs for hot rod images were calculated as relative values using 6 mm as the reference, according to the maximum value in the ROI for each hot rod. Inverse RCs of cold rod images were calculated using the following Eq. (3) [21]:

Inverse RC= \(\:1-\frac{{B}_{i}}{A}\) (i= 1–6)

where A represents the minimum counts in the ROI of the BG region inside the 1–6 mm cold rod parts, and B_i reflects the minimum counts in the ROI of the 1–6 mm cold rod parts. For uniformity evaluation, the pool section of the hot rod phantom was employed. A circular ROI covering 80% of the area was drawn on the transverse image of the pool section (Fig. 3). The percent coefficient of variation (%CV) was calculated from the average value and standard deviation in the ROI. The contrast, RC, and %CV were compared between ^99mTc and ¹⁸F.

The contrast of hot rod and cold rod images was shown in Fig. 4. The contrasts for the 1–6 mm hot rods were 1.1, 2.1, 2.8, 3.3, 3.6, and 3.8 for ^99mTc, and 1.0, 1.6, 2.4, 2.9, 3.2, and 3.3 for ¹⁸F, respectively. The contrast of the hot rod for ^99mTc and ¹⁸F demonstrated lower values with a decreasing rod diameter. Moreover, the ^99mTc image showed a higher contrast than the ¹⁸F image and approached the true contrast. The contrasts for 1–6 mm cold rod parts were 10%, 32%, 59%, 70%, 80%, and 86% for ^99mTc, and 0%, 7%, 28%, 49%, 63%, and 71% for ¹⁸F, respectively. The contrast in cold rod images with ^99mTc and ¹⁸F revealed a similar tendency as the hot rod contrast.

The RCs of 1–6 mm hot rods were 0.27, 0.51, 0.81, 0.98, 0.97, and 1.00 for ^99mTc and 0.28, 0.46, 0.76, 0.97, 1.05, and 1.00 for ¹⁸F, respectively (Fig. 5a). The RCs for the hot rods with 4–6 mm diameters were similar, whereas hot rods with diameters ≤ 3 mm demonstrated lower values as the rod diameter declined. Furthermore, the RC of the 5 mm hot rod in ¹⁸F was higher than that of the 6 mm hot rod. The inverse RCs of 1–6 mm cold rod parts were 0.06, 0.33, 0.68, 0.84, 0.93, and 0.96 for ^99mTc, and 0.00, 0.26, 0.53, 0.76, 0.83, and 0.94 for ¹⁸F, respectively (Fig. 5b). The inverse RC revealed lower values with a declining cold rod diameter. Moreover, the inverse RC for the cold rod image with ¹⁸F was lower than that with ^99mTc.

The %CV of the pool section for ^99mTc and ¹⁸F images was 4% and 7%, with the ^99mTc image displaying a lower %CV compared to the ¹⁸F image. Transverse images of the hot rod, cold rod, and pool sections were shown in Fig. 6. The hot rod and cold rod images with ^99mTc revealed a higher contrast compared with those with ¹⁸F. The pool image with ^99mTc appeared more homogeneous than that with ¹⁸F.

We evaluated the physical indices of SPECT and PET images utilizing the new phantom. Our phantom allows for the assessment of known contrast in hot rod images, as well as the contrast and RC of cold rod images, in addition to the physical indices assessable with the NEMA NU 4 phantom. The influence of PVE on the cold rod image is necessary to interpret the quantitative evaluation of the ischemic region [22, 23]. Furthermore, the overestimation or underestimation of the quantitative value in the tumor region may lead to PVE and Gibbs artifacts in nuclear oncology [24, 25].

While the NEMA NU 4 phantom is useful for verifying the accuracy of absolute quantitation using Bq/mL with a well counter or dose calibrator as a reference [26, 27], the accuracy of the absolute quantitative index is dependent on the dosimeter error, such as the dose calibrator and well-counter, as well as the variation of the cross-calibration factor (CCF) and errors caused by the reconstructed image [28–30]. Consequently, CCF uncertainty may arise from factors such as acquisition, image reconstruction, and other variables. Delineating between these factors will enhance quantitative accuracy. Relative evaluation in phantom experiments is excellent for detecting acquisition and image reconstruction errors in SPECT and PET systems. The contrast can be briefly set to a theoretical value by preparing solutions of two different radioactivity concentrations. The contrast ratio was set to 4, and the hot rod with a 6 mm diameter at ^99mTc obtained a contrast close to 4 in this study. The contrast of ¹⁸F was lower than that of ^99mTc owing to spatial resolution limitations. The spatial resolution of ^99mTc and ¹⁸F for VECTor⁺/CT has been reported to be better for ^99mTc [17]. High spatial resolution systems display a high contrast owing to the small PVE. Hence, the relationship of the hot rod contrast between ^99mTc and ¹⁸F reflected the previous study findings, and a similar trend was noted for cold rod contrasts.

The RC of the hot rod phantom remained relatively constant between 4–6 mm. Generally, PVE occurs for diameters < 2–3-fold the FWHM. The FWHM measured by the line source in VECTor⁺/CT is approximately 1.0 mm for both ^99mTc and ¹⁸F, aligning with the physical index [18]. Conversely, the RC was overestimated for the 5 mm diameter hot rod with ¹⁸F. The VECTor⁺/CT system applies a spatial resolution correction to ¹⁸F only. Although spatial resolution correction can improve the high-resolution image quality, it also produces Gibbs artifacts [25, 31]. The overestimation of the 5 mm diameter hot rod with ¹⁸F may be influenced by Gibbs artifacts.

The inverse RC of the cold rod phantom showed low values with a decreasing rod diameter, demonstrating a different behavior compared to the hot rod phantom. This disparity may be attributed to the presence of scattered radiation included in the cold rods. The ¹⁸F photons are strongly influenced by septal penetration from the collimator, as well as Compton scattering, due to their higher radiation energy compared with ^99mTc photons [32]. The uniformity of the pool section with ^99mTc and ¹⁸F also reflects these differences. Scattered radiation, septal penetration, and image noise are more pronounced in cold rod parts than in hot rod parts. Consequently, the inverse RC of ¹⁸F is predicted to be lower than that of ^99mTc.

In small animal studies, the radioactivity concentrations of ¹⁸F were higher than those typically encountered. PET images on the VECTor⁺/CT scanner were acquired using a SPECT system rather than a coincidence system. Moreover, the detector uses a NaI(Tl) scintillator with a 9.5 mm crystal thickness, resulting in counting losses for high-energy photons such as ¹⁸F. A previous study reported that only 10% of all ¹⁸F photons are detected as signals [17]. Hence, high-energy nuclides such as ¹⁸F require higher radioactivity concentrations.

The new phantom could evaluate PVE using RC due to its structure, which included large rods corresponding to the small animal SPECT and PET systems. Hot rods with RC values of 4–6 mm demonstrated similar values. This indicates that PVE is not considerably affected in rods larger than 4 mm under the acquisition and imaging conditions at our institution. In addition, the new phantom will facilitate the interpretation of quantitative evaluations of small animal SPECT and PET images, allowing the assessment of physical phenomena in both hot rods and cold rods.

We have developed a new phantom that allows physical phenomenon evaluation in small animal SPECT and PET images. This new phantom can assess the image contrast, RC, and uniformity of both hot rod and cold rod images. Additionally, it can evaluate the performance of small-animal SPECT and PET systems.

Acknowledgments

We would like to thank Akira E of Hokuriku E.P., Japan for his excellent technical support of phantom production.

Funding

This study was supported by JSPS KAKENHI Grant Number 18K15628.

Disclosure

The authors report no potential conflicts of interest relevant to this study.

Liu T, Wu Y, Shi L, Li L, Hu B, Wang Y, et al. Preclinical evaluation of [^99mTc]Tc-labeled anti-EpCAM nanobody for EpCAM receptor expression imaging by immuno-SPECT/CT. Eur J Nucl Med Mol Imaging. 2022;49(6):1810–21.
Jang BS. MicroSPECT and MicroPET Imaging of Small Animals for Drug Development. Toxicol Res. 2013;29(1):1–6.
Case JA, Bateman TM. Taking the perfect nuclear image: quality control, acquisition, and processing techniques for cardiac SPECT, PET, and hybrid imaging. J Nucl Cardiol. 2013;20(5):891–907.
National Electrical Manufacturers Association. NEMA, Standards Publication. NU 4–2008 performance measurements of small animal positron emission tomographs. Rosslyn: National Electrical Manufacturers Association; 2008.
Harteveld AA, Meeuwis AP, Disselhorst JA, Slump CH, Oyen WJ, Boerman OC, et al. Using the NEMA NU 4 PET image quality phantom in multipinhole small-animal SPECT. J Nucl Med. 2011;52(10):1646–53.
Teuho J, Han C, Riehakainen L, Honkaniemi A, Tirri M, Liljenbäck H, et al. NEMA NU 4-2008 and in vivo imaging performance of RAYCAN trans-PET/CT X5 small animal imaging system. Phys Med Biol. 2019;64(11):115014.
Gaitanis A, Kastis GA, Vlastou E, Bouziotis P, Verginis P, Anagnostopoulos CD. Investigation of Image Reconstruction Parameters of the Mediso nanoScan PC Small-Animal PET/CT Scanner for Two Different Positron Emitters Under NEMA NU 4-2008 Standards. Mol Imaging Biol. 2017;19(4):550–9.
Lajtos I, Czernin J, Dahlbom M, Daver F, Emri M, Farshchi-Heydari S, et al. Cold wall effect eliminating method to determine the contrast recovery coefficient for small animal PET scanners using the NEMA NU-4 image quality phantom. Phys Med Biol. 2014;59(11):2727–46.
Li S, Zhang J, Krol A, Schmidtlein CR, Feiglin D, Xu Y. Preconditioned alternating projection algorithm for solving the penalized-likelihood SPECT reconstruction problem. Phys Med. 2017;38:23–35.
Noori-Asl M, Sadremomtaz A, Bitarafan-Rajabi A. Evaluation of three scatter correction methods based on estimation of photopeak scatter spectrum in SPECT imaging: a simulation study. Phys Med. 2014;30(8):947–53.
Mok GS, Tsui BM, Beekman FJ. The effects of object activity distribution on multiplexing multi-pinhole SPECT. Phys Med Biol. 2011;56(8):2635–50.
Miwa K, Yoshii T, Wagatsuma K, Nezu S, Kamitaka Y, Yamao T, et al. Impact of γ factor in the penalty function of Bayesian penalized likelihood reconstruction (Q.Clear) to achieve high-resolution PET images. EJNMMI Phys. 2023;10(1):4.
Chen S, Gu Y, Yu H, Chen X, Cao T, Hu L, Shi H. NEMA NU2-2012 performance measurements of the United Imaging uPMR790: an integrated PET/MR system. Eur J Nucl Med Mol Imaging. 2021;48(6):1726–35.
Attarwala AA, Hardiansyah D, Romanó C, Jiménez-Franco LD, Roscher M, Wängler B, et al. Performance assessment of the ALBIRA II pre-clinical SPECT S102 system for ^99mTc imaging. Ann Nucl Med. 2021;35(1):111–20.
Attarwala AA, Karanja YW, Hardiansyah D, Romanó C, Roscher M, Wängler B, et al. Investigation of the imaging characteristics of the ALBIRA II small animal PET system for ¹⁸F, ⁶⁸Ga and ⁶⁴Cu. Z Med Phys. 2017;27(2):132–44.
Magota K, Kubo N, Kuge Y, Nishijima K, Zhao S, Tamaki N. Performance characterization of the Inveon preclinical small-animal PET/SPECT/CT system for multimodality imaging. Eur J Nucl Med Mol Imaging 38(4): 742–52.
Miwa K, Inubushi M, Takeuchi Y, Katafuchi T, Koizumi M, Saga T, et al. Performance characteristics of a novel clustered multi-pinhole technology for simultaneous high-resolution SPECT/PET. Ann Nucl Med. 2015;29(5):460–6.
Shibutani T, Onoguchi M, Kanno T, Ogihara S, Wakabayashi H, Inaki A, et al. Characteristics among three multi-pinhole collimators using a new small animal SPECT-PET/CT system. J Nucl Med. 2017;58(suppl 1):1124.
Branderhorst W, Vastenhouw B, Beekman FJ. Pixel-based subsets for rapid multi-pinhole SPECT reconstruction. Phys Med Biol. 2010;55(7):2023–34.
Ogawa K, Harata Y, Ichihara T, Kubo A, Hashimoto S. A practical method for position-dependent Compton-scattered correction in single photon emission CT. IEEE Trans Med Imaging. 1991;10(3):408–12.
Ichikawa H, Shibutani T, Shimada H, Okuda K, Kato T, Nosaka H, et al. Feasibility of using counts-per-volume approach with a new SPECT phantom to optimize the relationship between administered dose and acquisition time. Radiol Phys Technol. 2023;16(2):244–53.
Wakabayashi H, Taki J, Inaki A, Hiromasa T, Okuda K, Shibutani T, et al. Quantification of Myocardial Perfusion Defect Size in Rats: Comparison between Quantitative Perfusion SPECT and Autoradiography. Mol Imaging Biol. 2018;20(4):544–50.
Hess A, Nekolla SG, Meier M, Bengel FM, Thackeray JT. Accuracy of cardiac functional parameters measured from gated radionuclide myocardial perfusion imaging in mice. J Nucl Cardiol. 2020;27(4):1317–27.
Marquis H, Willowson KP, Bailey DL. Partial volume effect in SPECT & PET imaging and impact on radionuclide dosimetry estimates. Asia Ocean J Nucl Med Biol. 2023;11(1):44–54.
Rogasch JM, Hofheinz F, Lougovski A, Furth C, Ruf J, Großer OS, et al. The influence of different signal-to-background ratios on spatial resolution and F18-FDG-PET quantification using point spread function and time-of-flight reconstruction. EJNMMI Phys. 2014;1(1):12.
Iida H, Nakagawara J, Hayashida K, Fukushima K, Watabe H, Koshino K, et al. Multicenter evaluation of a standardized protocol for rest and acetazolamide cerebral blood flow assessment using a quantitative SPECT reconstruction program and split-dose ¹²³I-iodoamphetamine. J Nucl Med. 2010;51(10):1624–31.
Seo Y, Teo BK, Hadi M, Schreck C, Bacharach SL, Hasegawa BH. Quantitative accuracy of PET/CT for image-based kinetic analysis. Med Phys. 2008;35(7):3086–9.
Sanderson T, Solomon J, Nottage C, Dickson J. Underestimation of ⁶⁸Ga PET/CT SUV caused by activity overestimation using default calibrator settings. Phys Med. 2019;59:158–62.
Ichikawa H, Onoguchi M, Shibutani T, Kato T, Ito T, Shimada H. Optimization of cross-calibration factor for quantitative bone SPECT without attenuation and scatter correction in the lumbar spine: head-to-head comparison with attenuation and scatter correction. Nucl Med Commun. 2021;42(12):1404–16.
Miyaji N, Miwa K, Motegi K, Umeda T, Wagatsuma K, Fukai S, et al. Validation of Cross-calibration Schemes for Quantitative Bone SPECT/CT Using Different Sources under Various Geometric Conditions. Jpn J Radiological Technol. 2017;73(6):443–50. (in Japannese).
Hewitt E, Hewitt RE. The Gibbs-Wilbraham phenomenon: An episode in fourier analysis. Arch Hist Exact Sci. 1979;21:129–60.
Laymon CM, Turkington TG. Characterization of septal penetration in 511 keV SPECT. Nucl Med Commu. 2006;27(11):901–9.

Download PDF

Reviewers agreed at journal
28 Oct, 2024
Reviewers invited by journal
20 Aug, 2024
Editor assigned by journal
16 Aug, 2024
First submitted to journal
14 Aug, 2024

You are reading this latest preprint version

Development for a novel phantom for evaluating image quality in small-animal single photon emission computed tomography and positron emission tomography.

Status:

Version 1

Abstract

Objectives

Methods

Results

Conclusion

Figures

Introduction

Material and Methods

Phantom design

Phantom preparation

Acquisition and image reconstruction parameters

Image assessment

Results

Discussion

Conclusion

Declarations

References

Status:

Version 1