Click on an contribution preview the abstract content.
Click on an contribution preview the abstract content.
Development of a dual head PET system for nanoparticles nuclear imaging during magnetic hyperthermia treatment (#143)
M. Georgiou1, E. Fysikopoulos1, E. Lamprou2, A. Gonzalez-Montoro2, A. J. Gonzalez2, G. Loudos1
1 Technological Educational Institute of Athens (TEIA), Department of Biomedical Engineering, Egaleo, Greece
Nanoparticle based drug delivery is considered as a promising technology for the efficient treatment of various diseases. Use of magnetic nanoparticles (NPs) in hyperthermia treatment is one of the most promising directions. Simultaneous nuclear imaging during hyperthermia can provide insights in the biological process that occur when NPs are heated and thus monitor the successful organ targeting, drug release and/or real time response to therapy. However, placing an imaging system inside or very close to magnetic field has many technical challenges and such a system doesn’t exist yet.
The project THERAQ develops and validates procedures and support for Quality Assured Theranostics and has been recently funded by the Greek Scholarships Foundation. To overcome the technical challenges, SiPM detectors, already used in the emerging field of PET/MRI are exploited. The main objectives of THERAQ are (Figure 1) the (i) Development of a PET prototype system for use inside the coil of hyperthermia systems; (ii) Design of a hyperthermia coil, in order to be combined with the PET system for in vivo imaging; (iii) Differentiation of active and passive targeting through the spatial distribution of the pharmaceutical measured in tumor and (iv) Imaging of drug release during hyperthermia, using radiolabelled pharmaceutical or radioisotope entrapped inside nanoparticles.
Two PET detectors with MR compatible capabilities have been designed and used throughout all the experiments. Each PET detector block is composed by a BGO crystal array with 22×22 elements (2 mm size, 5 mm height) coupled to a 12×12 SiPM array with 3×3mm2 active area each. Custom flex PCB boards 40cm long, were developed to fed up the signals coming from the readout electronics, avoiding additional cabling and connectors inside the magnetic field. Each readout circuit provides an analog signal for each row and column of the SiPM array. The 24 signals and their sum are connected to a high performance custom DAQ system using coaxial flat cables. The DAQ is composed by a trigger board that accepts the events based on predefined analog thresholds. Two ADC boards work in parallel with up to 66 channels. The events are being processed and packaged inside a FPGA. Concerning hyperthermia system, two RF coils have been constructed for use with Ambrell’s Easyheat induction heating system.
The added value of THERAQ can be summarized as Technical improvements in hyperthermia and temperature measurement equipment; Construction and evaluation of the first Theranostic imager for hyperthermia and New protocols for quality assured Theranostics.
 Ahmed N, Fessi H and Elaissari A 2012 Drug Discovery Today 17 928
 Hayashi K, Nakamura M, Miki H, Ozaki S, Abe M, Matsumoto T, Sakamoto W, Yogo T and Ishimura K 2014 Theranostics 4 834]
 A.J. González, et al., “The MINDView brain PET detector, feasibility study based on SiPM arrays,” Nucl. Instr. Meth. Phys. Res. A, vol. 818, pp 82-90, 2016.
 EASYHEAT 1.2 to 2.4 kW Induction Heating Systems, Ambrell, available online: pdfo.ambrell.com/411-0050-10.pdf
This research is carried out in the frame of the COST Action TD1401 supported by COST (European Cooperation in Science and Technology) and through IKY scholarships programme and co-financed by the European Union (European Social Fund - ESF) and Greek national funds through the action entitled ”Reinforcement of Postdoctoral Researchers”, in the framework of the Operational Programme ”Human Resources Development Program, Education and Lifelong Learning” of the National Strategic Reference Framework (NSRF) 2014 – 2020.
“β-eye”: A low cost, portable coincidence camera for whole-body mouse dynamic imaging (#101)
M. Georgiou1, E. Fysikopoulos1, G. Loudos1, 2, 3
1 BioEmission Technology Solutions, R&D, Athens, Greece
Animal models are essential in the development of new imaging agents, genetic research and pathophysiologic investigations and drug development safety , . Several multimodal imaging systems are commercially available for preclinical research, but their purchase and maintenance costs make them non affordable for the majority of small and medium research groups. Taking into account the average end user needs and following the development of our scintigraphic system “γ-eye”, we present “β-eye”, a whole body mouse coincidence camera suitable for in-vivo molecular imaging of PET isotopes.
Each “β-eye” detector consists of a 22×44 BGO scintillator array with crystal elements of size 2×2×5 mm3, coupled to two H12700A PSPMTs , covering a field of view of 50×100 mm2, large enough for the majority of laboratory mice. Two of the aforementioned detectors are placed one physically opposed to the other at a distance of 80 mm, forming a dual-head geometry optimized for imaging mice. The detectors are kept stationary during the scan, making “β-eye” a limited angle tomography system, suitable for all PET isotopes. The external dimensions of the entire system are 30×32×35 mm3 (Figure 1 (left)). “β-eye” is accompanied with an appealing software providing user defined real time imaging and easy-handled tools for post processing via an intuitive database archive (Figure 1 (right)).
Performance characteristics of “β-eye” were determined experimentally using 18F-FDG, which is the most common radiotracer for PET applications. State of the art methodology was followed in all measurements . The spatial resolution of the system is 3.2 at the center of the detectors. The energy resolution is 26% at 511 keV and the maximum recorded sensitivity 14 kcps/MBq. The measured timing resolution is equal to 8.2 nsec. “β-eye” system was evaluated in a proof-of-concept animal study using normal Webster Swiss Albino mice with weight 25 gr. The mouse was injected with 100 ul/ 1.7 MBq 18F-FDG and static images were obtained at 10 min p.i. and at 60 min p.i. (Figure 2). The biodistribution of the tracer is clearly given and several intermediate time points can be derived.
The presented results demonstrate the ability of “β-eye” to be used as an efficient standard screening tool for daily research supporting the imaging needs of research groups that have access to radioactivity, inject animals with radiolabeled compounds, but then use ex vivo biodistributions. While multimodal systems can fully validate a new probe, performance of “β-eye” supports the argument that coincidence imaging of PET isotopes can provide in a cost effective and accurate way, a very good indication of the in vivo biodistribution boosting research in drugs and biomolecules development.
 Peterson TE and Shokouhi S (2012) Advances in preclinical SPECT instrumentation. J Nucl Med 53:841–844.
 Gomes CM, Abrunhosa AJ, Ramos P and Pauwels EKJ (2011) Molecular imaging with SPECT as a tool for drug development. Adv Drug Deliv Rev 63:547–554.
 Hamamatsu Photonics, “Flat panel type multianode pmt assembly H12700 series.” 2014.
 Zhang H et al. (2011) Performance Evaluation of PETbox: A Low Cost Bench Top Preclinical PET Scanner. Mol. Imaging Biol. 13:949–961.
This research has been co-financed by the Greek national funds through the National Strategic Reference Framework (NSRF) – 2014-2020, code: Τ1ΕΔΚ-01159 with acronym NAVIGATE.
Keywords: PET, nuclear imaging, in vivo imaging, preclinical research
A novel PET scanner based on an edgeless, continuous scintillator tube (#324)
S. Berr1, A. J. Gonzalez2, S. Majewski1, M. Williams1, A. Gonzalez-Montoro2, G. Cañizares2, F. Sanchez2, J. M. Benlloch2
1 University of Virginia, Charlottesville, Virginia, United States of America
Most commercially available PET are based on pixelated technology. As an alternative, monolithic scintillators have been employed and photon absorption location is calculated with high precision from the shape of the scintillation light distribution. 3D absorption localization for both pixelated and continuous crystals is inaccurate at the edges because the light pattern is truncated, leading to a degradation of the spatial resolution. Spatial resolution will improve if the edges of the scintillator are removed. Using a single crystal scintillator tube would effectively remove all edges.
We acquired a coincidence data set using standard continuous scintillators and compared the resolution in the reconstructed image with that obtained using the same data but omitting coincidences with one or both photons near a crystal edge. Images were reconstructed using lines of response (LORs) that included the entire crystals (Original) and with LORs involving only the inner 60% of each crystal (Filtered), simulating the proposed continuous-tube detector behavior. The data set was acquired using a mini Derenzo phantom (rods starting at 0.75 mm) imaged with a prototype Bruker Albira Si small animal PET insert. The insert is currently installed at University of Leuven, Belgium.
Images resulting from these reconstructions are shown in the Figure. We have reconstructed the data using MLEM algorithm with 35 iterations, voxel and virtual pixel sizes of 0.25 mm and 1.5 mm, respectively. This scanner employs continuous crystals (10 mm thickness), which are suitable to simulate an ideal detector without edge effects. DOI is determined by the shape of the scintillation light pattern. Projections through the hot rods were obtained and fitted with Gaussians. The fitted Gaussian width (which is proportional to FWHM) is 1.45 mm on average for the Original data set and 1.19 mm for the Filtered data set. This represents a roughly 23% increase in resolution. The measured spatial resolution for the Albira Si insert is 0.75 mm. Based on these results, the resolution of the edgeless detector system is estimated to be 0.6 mm.
Before constructing a PET scanner based on the scintillation tube design, we wanted to devise a test that would demonstrate the advantages of the edgeless scintillator design. We have done that by removing edge data post-acquisition using data from an existing scanner. A PET scanner built around an ‘edgeless’ scintillator tube design is expected to have a spatial resolution of 0.6 mm, which approaches the spatial resolution limits of PET.
(1) Gonzalez A, et al. Trans Nuc Sci 63, 2471, 2016;
(2) Gonzalez A, et. IEEE Med Imag Conf 2015;
(3) Pani R, et al. IEEE Trans Nuc Sci 63, 2487, 2016;
(4) Sanchez F, et al. Med Phys 39, 643, 2012.
Shown are projections through a mini Derenzo phantom obtained on an Albira Si scanner using all the data (Original) and using data only from the center of the scintillators (Filtered). The projections are fitted with a Gaussian (solid lines). The fitted Gaussian width (which is proportional to FWHM) is about 25% narrower for the filtered case.
Keywords: PET insert, Monolithic crystals
High throughput PET/CT imaging using a multiple mouse imaging system (#132)
H. E. Greenwood1, Z. Nyitrai2, G. Mocsai2, S. Hobor2, T. H. Witney1
1 University College London, Centre fo Advanced Biomedical Imaging, London, United Kingdom
A considerable limitation of current small animal PET/CT imaging is the low throughput of image acquisitions which becomes particularly restrictive when radioactive isotopes with short half-lives, such as Fluorine-18, or complex dynamic imaging studies are employed. Subsequently, to design sufficiently-powered studies, high costs accumulate. Together with Mediso Medical Imaging Systems, a four bed mouse ‘hotel’ was developed to simultaneously image up to four mice (Fig 1A), thereby reducing the cost and maximising radiotracer usage when compared to scans performed with a single mouse bed.
The bed was initially evaluated using four mini image quality (IQ) phantoms filled with ~3.7MBq of [18F]fluorodeoxyglucose ([18F]FDG) imaged over 20 min by PET/CT (Mediso NanoScan; 1-5 coincidence mode; CT attenuation-corrected; scatter corrected). Post reconstruction, NEMA tests for uniformity, recovery coefficients (RC) and spill-over ratios (SOR) were performed (Tera-Tomo 3D; 6 iterations, 6 subsets). Heat and anaesthesia tests were carried out on the bed using a thermal camera (FLIR systems AB- E60) and flowmeter (Dwyer RMA-26-68V), respectively. Subsequently, four mice bearing HCT116 KRAS mutant colorectal tumour xenografts (~100mm3) were simultaneously imaged by dynamic PET over 60 min following the intravenous injection of ~3.7MBq [18F]FDG.
Following image acquisition, little variance in recovered activity between the simultaneously-imaged phantoms was seen (Fig 1C). Mediso NEMA tests performed on the reconstructed mini IQ phantom images resulted in uniformity and RC values for the 5 mm diameter rods to be within the tolerable limits (10.3%±0.4% and 1.2±0.1, respectively; Fig 1D i), with SOR in the non-radioactive water and air filled chambers also within tolerable limits (0.1± 0.02% and 0.1± 0.02%, respectively; Fig 1D ii). Temperature remained constant throughout the bed (36.8°C±0.4°C, n=4) and anaesthesia distributed evenly to each nose cone (2.9±0.1 L/min, n=4). To evaluate the bed under real-world conditions, four tumour-bearing mice were simultaneously imaged over 60 min (Fig. 2A). Low variability in radiotracer uptake was observed for all major organs (Fig 2B i-iii). Variation in [18F]FDG tumour uptake, however, was observed; thought to reflect differences in metabolic activity between the tumours (Fig. 2B iv).
The multiple mouse imaging system was designed and manufactured for uniform control over temperature and anaesthesia, with space for up to four mice within the same PET/CT field of view (Mediso NanoScan). Analysis of images acquired using the four bed animal hotel confirmed its utility to increase the throughput of small animal PET imaging without the loss of image quality and quantitative precision. In comparison to a single mouse bed, cost and time associated with each scan were substantially reduced.
Fig 1. Phantom PET/CT image acquisition using the four bed mouse hotel.
A) The four bed mouse hotel. The modular design with removable top bed allows for 1-4 animals to be simultaneously imaged. B) 20 min PET/CT image acquisition of mini IQ phantoms containing ~3.7 MBq each. C) Phantom TAC showing low variability between phantoms imaged simultaneously. Shaded regions represent one s.d. from the mean value (n = 4). D) NEMA test results for uniformity (i) and SOR (ii).
Fig 2. Dynamic PET imaging of tumour-bearing mice using the mouse hotel.
A) Representative 40-60 min summed activity of [18F]FDG uptake in four HCT116 KRAS mutant tumour-bearing mice. Dashed lines indicate the tumours. B) Brain (i), liver (ii), kidney (iii) and tumour (iv) TACs, normalised to the percentage injected activity. Shaded regions represent one s.d. from the mean value (n = 5 animals).
Keywords: high throughput, PET imaging, FDG
Design and performance study of quasi-spherical PET scanner (#194)
D. Perez-Benito1, J. M. Udías2, M. Desco Menéndez1, 3, 4, J. J. Vaquero1, 3
1 Universidad Carlos III de Madrid, Departamento de Bioingeniería e Ingeniería Aeroespacial, Leganes, Madrid, Spain
Relevant challenges on new Positron Emission Tomography (PET) scanner design face are, besides their compactness and cost, high sensitivity, spatial and temporal resolution to quantify biological processes. Significant physical effects on the resolution are being assessed; namely, positron range, photon acollinearity, detector Point Spread Function (PSF) and sensitivity. The sensitivity is presented versus the dose as the Noise Equivalent Count rate (NEC). In this work, we present a Proof of Concept (PoC) design and its figures of merit for a preclinical PET scanner shaped as an icosahedron.
The PoC icosahedron is being built with 20 facets of 155mm of edge (inscribed sphere of 234mm diameter) covered by 10 hexagonal scintillator crystals (Fig. 1) . For human brain studies, the facets would be extended with five crystals for a total facet length of 194mm. Depth of Interaction (DoI) information is obtained by engraving different layers in the crystal scintillator by means of Sub-Surface Laser Engraving (SSLE) . The NEC curve has been simulated with a uniform distribution of 18F in a sphere of water of 2cm of radius at the center of the FOV, dead time of 200ns with 100ns pile-up, 2ns coincidence time and 350-650keV energy window. Resolution is studied with a point source of 18F at various locations of the FOV according to NEMA protocols. Simulations are done with GATE 7.2.
Coincidence lines are stored in a three-dimensional histogram. Fig. 2 shows a projection of the angular coverage (azimuthal and zenith angles) for a γ-point source placed at the center of the FOV. It represents the range of visible angles with blind spots at the junction of the faces and fewer detected lines of response at the sides of the faces. As expected, the area with higher sensitivity is the center of the facets where it is in direct coincidence with another facet in diametrical opposite direction. The NEC is presented as an estimation of the true, random and scatter event rates for a given activity distribution. It is a measure of the expected signal-to-noise ratio. The high sensitivity due to the large solid angle coverage pushes the NEC knee far beyond typical scanners. The estimated sensitivity is about 18% for an energy window of 350-650keV and a LSYO scintillator crystal 10mm thick. This arrangement results in a geometrical efficiency of 73% with respect to a 4π coverage.
This design has favorable characteristics for fast, dynamic high-resolution PET imaging. Preliminary results confirm our initial hypothesis about the high-resolution, high-sensitivity of the design, as well as the stringent requirements for the readout of the SiPM arrays channels. The high sensitivity would favor low dose experiments or very high time resolution on dynamic scans. Both options are highly desirable in a translational version of this scanner as an alternative to helmet-like designs for brain PET imaging.
 D. Perez-Benito, R. Herrera, R.Chil, G.Konstantinou, J.M.Udías, M.Desco, J.J. Vaquero. (2017, Nov). Proposal for a PET scanner with 4π steradian span. Conference Records presented at the IEEE Nuclear Science Symposium and Medical Imaging Conference, Atlanta.
 G. Konstantinou, R. Chil, M. Desco, and J. J. Vaquero, “Sub-Surface Laser Engraving Techniques for Scintillator Crystals: Methods, Applications and Advantages,” IEEE Trans. Radiat. Plasma Med. Sci., pp. 1–1, 2017, DOI: 10.1109/TRPMS.2017.2714265.
This work was partially funded by the Human Frontier Science Program (RGP0004/2013), projects RTC-2015-3772-1 and TEC2016-78052-R from the Spanish Ministerio de Ciencia e Innovación, and project TOPUS S2013/MIT-3024 from the regional government of Madrid. The research leading to these results has received funding from the Innovative Medicines Initiative (www.imi.europa.eu) Joint Undertaking under grant agreement n°115337, resources of which are composed of financial contribution from the European Union's Seventh Framework Programme (FP7/2007-2013) and EFPIA companies' in kind contribution. This work also acknowledges support by EU’s H2020 under MediNet, a Networking Activity of ENSAR-2 (grant agreement 654002).
Preclinical system PoC scaled from a clinical head scanner design, placed on a flat support. The scanner is formed by twenty modules arranged on ten sectors of two facets, half of them supported by hinges (orange); those sectors are moved by electromechanical actuators (not shown) to open the scanner. In order to ease the presentation, the CB-X ray tubes are not included either.
Keywords: Positron Emission Tomography (PET), Biomedical imaging, Scanner design, Brain imaging
PET-CT to investigate regional lung deposition of aerosols in rodents: comparison of three nebulisation systems (#161)
U. Cossío1, V. Gómez-Vallejo1, M. Flores2, J. Llop1
1 CIC biomaGUNE, Radiochemistry and Nuclear Imaging Goup, San Sebastian, Spain
Pulmonary administration of drugs has recently gained attention because it exhibits numerous advantages compared to oral or intravenous administration. The administration of aerosols for inhalation to animals, however, remains a critical challenge and only a few methods of administration have been developed. Herein, we compare the regional distribution of 18F-fluorodeoxyglucose ([18F]FDG) in the lungs of WT rats after pulmonary administration using three different inhalation methods.
[18F]FDG was kindly donated by Curium Pharma and was used as received. Aerosolized [18F]FDG was administered to anesthetized rats using three different administration methods: (a) The Penn-Century MicroSprayer® Aerosolizer; (b) an in-house designed aerosol generation system developed at Ingeniatrics Tecnologías; and (c) The Aeroneb® Lab Micropump Nebulizer. Dynamic PET images were acquired immediately after finalising the administration, using an eXplore Vista PET-CT system. Images were reconstructed by filtered back projection and analyzed using NIH Image-J processing software. Regional distribution of the aerosol in the lungs was determined by VOI analysis and local/regional clearance was assessed using voxel-by-voxel analysis.
Intratracheal nebulisation using the PennCentury MicroSprayer resulted in >80% of the administered dose accumulated in the lungs, with a non-uniform distribution of the radioactivity in different lobes (Figure 1a) and a low animal-to-animal reproducibility. Administration using either the aerosol generation system or the Aeroneb nebulizer resulted in a uniform distribution over the lungs (Figure 1b for example of distribution using the aerosol inhalation system developed by Ingeniatrics), but only a small fraction of the nebulized activity was deposited in the lungs. Uniform clearance from the lungs was observed irrespective of the administration method and initial distribution of the radioactivity.
Aerosol inhalation using both nebulizers results in a uniform distribution, while intratracheal nebulization leads to almost quantitative, although non-uniform, deposition of the aerosol within the lungs. [18F]FDG is a suitable tool for the assessment of regional distribution of the aerosol in the lungs following pulmonary administration.
This project has received funding from the European Union’s Seventh Framework Programme for research, technological development and demonstration under grant agreement no 604434.
Keywords: Lung Administration, [18F]FDG, Nebulisers, PennCentury MicroSprayer
Non-Invasive bone healing monitoring through SPECT/CT Imaging in Femur Bone Defect Mouse Models (#246)
M. Rouchota1, E. Fragogeorgi2, S. Sarpaki2, J. Daich3, M. Georgiou1, P. Bouziotis2, G. Loudos1, 2, 4
1 BET Solutions, Athens, Greece
Non-invasive monitoring of bone defect healing through imaging, gains interest in the evaluation of novel synthetic scaffolds in bone tissue engineering, as an alternative approach to the clinical gold standard treatment (autografting) [1,2]. Ιn the frame of the present study, we establish a SPECT and CT based evaluation procedure, for the longitudinal assessment of bone healing. This is a method that can be employed in the early stage of the in vivo evaluation of bone regeneration, exploiting the added value of molecular imaging information.
During the first step of this study, 2 mouse groups (n=5) were purchased from the Breeding Facilities of the NCSR “Demokritos” (Athens, Greece). A cortical bone defect, 0.8 mm in diameter, was created on the mid-diaphysis of the femur of each mouse. The gap was initially measured with a caliper and then monitored through CT and SPECT imaging for a period of 6 weeks. For the first group (control), empty defects were created, whereas for the second group defects were filled with collagen-based scaffold materials. For SPECT imaging, mice were injected with 99mTc-MDP two hours before imaging. The defect size was measured directly through the CT images and through the recorded counts in the defect area, as a percentage of counts over a healthy bone area, for SPECT imaging.
The exact size of the bone defects created on both groups was monitored through SPECT and CT imaging, for a total period of 6 weeks. For the first group (control), the empty defects (untreated mice) were evaluated before the surgical procedure, by being intravenously injected with 99mTc-MDP (Fig.1) and scanned at 2 h post-injection, following the surgical procedure (Fig. 2). The same procedure was followed for the mice where collagen-based scaffold materials were used and the difference induced by the material was quantified and analyzed, for all studied time points. The correlation of the two imaging methods provides more accurate and early information on the treatment of surgically-induced defects.
In vivo SPECT/CT imaging allowed for a detailed evaluation and assessment of the bone defect healing process in the femur of adult mice, with and without a therapeutic scheme. The efficiency of collagen-based scaffold materials was evaluated, regarding their ability to aid the bone healing process. Further analysis of the results could provide insights into the underlying mechanisms.
This research is carried out in terms of the H2020-MSC-RISE project VIVOIMAG (No. 645757) and is financially supported by the program of Industrial Scholarships of Stavros Niarchos Foundation (ISN), Greece.