OVERVIEW

Medical Imaging (MI) has affected diagnostic accuracy in diverging fields of Medicine. Despite the on-going progress of MI, it is still an open issue to achieve low-cost designs with enhanced image quality and reduced human burden. Computational techniques support innovative MI design especially for modalities utilising phosphors/scintillators, viz., materials that emit light when excited by radiation. There is considerable interest nowadays onflat-panel x-ray imaging detectors. Novel technologies are related to new materials in conjunction with state-of-the art optical sensors Interesting approaches concern (I) investigating new nanophosphors with enhanced characteristics and (II) modern Complementary Metal Oxide Semiconductor (CMOS) optical-sensor technology. Any new approach prerequisites intensive research prior to adoption in commercial MI systems.

The research aims in (a) investigating new phosphor and nanophosphor detectors for modern MI systems in combination with different optical sensors, and, (b) developing computational Monte Carlo (MC) methods targeted to the applicability of phosphor and nanophosphor detectors in MI. The outcomesenable (1) the suggestion of optimum phosphor detector-optical sensor combinations under various conditions, and, (2) suggestion of newMI detector designs especially with nanophosphors coupled to CMOS sensors. Whole activity is expected to affect (i) industrial MI applications, and (ii) research of biomedical nanomaterial technology. These will benefit MI industry, health care services, clinical applications and research.

INTRODUCTION

Non invasive MI techniques have developed widely. Related diagnosis refers to structural, functional and physiological processes. Principal branches are (a) transmission and (b) emission imaging. First branch refers to x-rays, transmitted through human body or, diffracted by macromolecules. Characteristic applications are diagnostic radiology, x-ray radiography, fluoroscopy, angiography, computed tomography (CT), portal imaging and protein crystallography. Second branch includes two large categories, namely Single Emission Computed Tomography (SPECT) and Positron Emission Tomography (PET). SPECT and PET utilise radiation emitted by nuclides administered in human body. Administered nuclides trace biochemical and physiological processes as they label critical substrates, ligands, drugs, antibodies, neurotransmitters and other compounds. Through this way functional and physiological information are provided, especially regarding cancer tumours.

Intensive research is implemented in enhancing characteristics of MI systems Special is the concern on investigating new-novel fluorescent materials and nanomaterials emphasising on the optimisation of new detector configurations. Fluorescent materials are widely used in several MI detectors as converters of x-rays to light (van Eijk 2002, Nikl 2006). They include scintillators, phosphors and nanophosphors. They are employed in form of screens or crystals, coupled to optical sensors, as e.g. CCDs, photodiodes and flat panels. Alternative detector configurations utilise photoconductors, which convert x-rays to charge. All types provide the electric signals that are necessary for image formation (van Eijk 2002). Phosphors and scintillators are discriminated according to the time needed for the emission of light. Nevertheless, in MI this time is incorporated in material's properties. Therefore MI phosphors and scintillators are referred altogether. For detection, scintillators are incorporated in suitable designs (van Eijk 2002).The spectral emission properties are important in selecting the appropriate fluorescent material .For example, the distribution of wavelengths of scintillation light is critical for matching a scintillator with available optical sensors. New promising materials and detector geometries have shown very interesting results both in enhancing detector performance and reducing overall cost; the latter rendering systems attractive for commercial production.

Especially regarding emission imaging, considerable is the effort towards non-invasive, high-resolution, small animal imaging technologies. Nonetheless, significant challenges remain to be overcome when, for example, attempting to image a 30-g mouse as compared with a 70-kg human. It is however still difficult to include enough spatial information to obtain meaningful anatomical and functional data. Multiple imaging modalities are available for small-animal molecular imaging including micro-PET, micro-CT, micro-SPECT. State-of-the art instrumentation uses advances in detector technology, including exploitation of solid-state detector technology and modern image-reconstruction techniques. The investigation in this field renders newer generations of scanners of high resolution, sensitivity, and sthroughput time. On the other hand, standard SPECT cameras still use NaI(Tl) single crystals. However, there are some prototypes of small field of view, where other crystals have been tested. Characteristic examples are the YAP, BGO, BaF and CsI(Tl) scintillators. In addition, some detectors employed solid-state detectors-usually CdZnTe- with excellent results. PET cameras are based mainly on BGO, GSO and LSO single crystals. Other scintillators are also proposed for PET prototypes. Mixed scintillators designs, the so called phoswitch detectors, are also under investigation.

Nowadays, special is the interest on flat panels. These are large area x-ray detectors that are position sensitive, i.e., they provide electric signals depending on the interaction site. Due to this fact they exhibit high image quality. For example, in x-ray projection imaging flat panel arrays with phosphor screens or photoconductors, detect attenuation profiles of human body's parts and produce accurate images. Mammography utilises thin phosphor screens or photoconductors which slot scanning devices based on phosphor-CCD detectors (van Eijk 2002). Chest Radiography employes large area flat panel arrays with thick phosphor screens. Fluoroscopy uses flat panels or image intensifiers with phosphor screens of fast response. In this manner, fast real-time images are produced (van Eijk 2002). In CT, ceramic scintillators or crystals are used, while portal imaging employes flat panels of very thick phosphor screens.

It becomes evident that scintillators constitute critical structural parts of detectors of MI systems. International research in the field of MI is focused on optimising performance of detectors involved. Investigations refer to optimization of detectors' geometry, material- for example, type of crystals, collimator and position-sensitive photomultiplier tubes-, electronics readout, data acquisition and image reconstruction. Since detector design affects resolution and sensitivity of the total system, many groups and companies evaluate a number of new materials for specific applications. It is this fact, that renews the necessity for investigating and evaluating new phosphors and nanophosphors for MI. Optimisation of existing MI materials can be achieved by improving their structural elements or image characteristics. Improvement in structure refers to thickness, grain size and columnar structure. Improvement in characteristics refers to enhancement of absorption and intrinsic efficiency and by providing better optical properties. Apart, brandnew designs can be suggested from modelling or experimental results. In addition to phosphors, novel CMOS technologies are under investigation to develop high resolution and low noise optical sensors with “system on chip” properties(Arvanitis et al 2009)In conclusion, further research is still needed in the field of detector optimisation.

Powder scintillator materials have been previously assessed in many fields of diagnostic radiology, namely in general radiography, mammography and computed tomograph. Additionally several single crystal scintillators have been studied for gamma-ray imaging. However, restrictions exist because (1) very limited number of studies are concerned with powder scintillators that might be employed in ring type SPECT detectors; (2) the response of powder scintillators has not been extensively studied in a wide range coating thickness values nor at exposure conditions corresponding to nuclear medicine modalities; (3) only restricted output image properties have been extensively studied, namely the spatial resolution, luminescence emission efficiency (LE), noise, and contrast; (4) physical properties such as the spectral compatibility and intrinsic optical properties like the optical absorption and scattering coefficients, light reflectivity, intrinsic conversion efficiency etc. have not been adequately investigated. Under investigation is also their relationship to image quality; (5) limited number of single crystal scintillators have been extensively investigated, principally those employed in commercial systems. Still, crystal treatment, on micro level, can possibly improve the performance of already existing materials.

Currently, powder scintillators are mainly ceramic phosphor materialsunder reduced porosity. Phosphor grains are glued through binding material in close packed spatial distribution (Nagarkar et al 2004; Blasse and Grabmeier 1994).Similarstructures are also addressed in nanophosphors. Ceramic phosphors and nanophosphors provide high detection efficiency and high image quality(Zych et al 2003, Liaparinos and Kandarakis 2009). Intrinsic and physical imaging properties of powder scintillators have been investigated through experimental and theoreticalmethods (Yaffe and Rowlands 1997), however for imaging systems of previous decades. To-date performance of detectors is evaluated in space andspatial frequency domains through frequency dependent parameters as e.g. Modulation Transfer Function (MTF), Noise Power Spectrum (NPS) and Detective Quantum Efficiency (DQE). MTF is measured frequently using the slanted-edge method to avoid aliasing while the Normalized NPS (NNPS) is determined by two-dimensional (2D) Fourier transforming of uniformly exposed images. Both measurements are performed under the representative radiation quality (RQA) settings, RQA-5 (70kVp digital-radiography) and RQA-M2 (28kVp digital-mammography) recommended by the International Electrotechnical Commission Reports 62220-1 and 62220-1-2 respectively. Detective Quantum Efficiency (DQE) can be assessed from the measured MTF, NPS and the entrance surface air-Kerma (ESAK) obtained from X-ray spectra measurement, normally with cadmium telluride (CdTe) detector.(IEC, 2005; Michail et al. 2011).

Scintillators of MIhave been successfully modelled through Monte Carlo methods. These were proven to be by far the most successful technique for the simulation of the stochastic processes involved in radiation detection (Nikolopoulos et al., 2006). Monte Carlo techniques have been successfully applied to medical physics, and particularly in evaluating MI detectors. Monte Carlo methods constitute, in principle, numerical-statistical approximations. Key-success, is the quantification of random processes in physical phenomena that are difficult, or even impossible, to determine experimentally.During the last decade, various Monte Carlo simulation packages have become commercially available. Some research groups have reported results on application of such packages in studies of photon transport phenomena in scintillators employed in x-ray medical imaging (e.g Nikolopoulos et al, 2006; Liaparinos and Kandarakis 2009 and references therein). However, commercially available Monte Carlo simulation packages are general and for this reason, their application is constrained by their expediency and feasibility in specialising to firm situations. The most popular packages i.e. EGSnrcMP, GEANT4 and PENELOPE, have been developed and verified for studies mainly in the field of nuclear and high energy physics (Nikolopoulos et al., 2006). Nevertheless, even such platforms may aid in finding significant information for effects within material's structure and the intrinsic properties of MI scintillators. In this sense, Monte Carlo simulation is very useful for complex problems that cannot be modeled by computer codes using deterministic methods (Badano and Sempau 2006). Results obtained by the Monte Carlo simulations can contribute to x-ray detector optimisation, toward improving whole imaging processes. This can be achieved by simulating (a) new efficient MI detectors, (b) novel advances in MI acquisition in conjunction with (c) lower patient doses.

OBJECTIVES

The objectives of the present research are:

1. The evaluation of the performance of powder scintillator radiation detectors under irradiation with various gamma-ray sources. Powder radiation scintillators are used in the form of thin layers (i.e. phosphor screens), which are prepared in laboratory. Various layers from various materials and with various thicknesses are prepared and experimentally studied. Under these conditions a number of physical quantities, related to the light emission and imaging performance of scintillators, are experimentally determined for various photon energies and scintillator detector thickness values. These quantities are as follows: a) The spatial resolution, b) The energy resolution, c) The detector sensitivity-expressed by the emitted light per unit of radiation exposure (uminescence emission efficiency), d) The light emission spectrum and the spectral compatibility to various optical sensors, e) The light transparency and the attenuation of the generated light (light losses) within the scintillator mass and f) The intrinsic light yield (conversion of the absorbed radiation energy into light within the scintillator material). In addition the scintillator behaviour will be modeled using theoretical models describing radiation and light transport within the scintillator mass. Modelling will allow for prediction of scintillator performance under conditions different to those experimentally available in our laboratories (e.g. very high or very low photon energies, very thick layers etc).

2. The comparison of the performance of powder vs. single crystal scintillators under identical irradiation conditions. This comparison is expected to provide useful data concerning possible applications of powder scintillators, especially in ring type SPECT detectors. Since powder scintillators are much cheaper and easier to manufacture - when compared to single crystals - the possibility of obtaining similar results between these to alternative scintillator forms will be of a great significance, especially in the design and implementation of low cost imaging systems. Such ring type detector systems already exist (microSPECT and microPET), but their cost do not allow their use in a number of dedicated imaging applications. Thus, the minimization of their cost is one of the main goals of this project.

3. The evaluation of the imaging properties of crystals in single form, after crystal treatment on micro level, using a high-resolution gamma camera. The surface and the exit window of the scintillator affect the light output and its optical properties. Detailed signal analysis will be carried out and phantom tests will be performed. The possibility of optimizing the performance of commercially available scintillators by performing an additional low cost entrance or output layer treatment and edge processing on micro-level may provide significantly important results. Recent studies have shown that the continuous crystals, used today in Nuclear Medicine, can be further processed and their performance can be further improved. The most promising geometries will be studied, since improvement of detectors performance on the crystal level is considered of high importance and may lead to significant innovations, with commercial applications.

4. The development of well proofed Monte Carlo simulation packages of the optical properties of single crystals and powder screens. Monte Carlo simulation packages and custom simulation codes will be employed and the experimental results will be used to evaluate the Monte Carlo simulation models. Original simulation models will be developed and evaluated, which will be further used in future studies. Since simulation is an important tool in the study of scintillator materials, the implementation of a well-tested simulation code, is scientifically very important.

Expected Results

The results expected from the proposed project allow for the following:

1. The estimation of the thickness and the type of a powder scintillator layer with emission efficiency and imaging properties approximating the efficiency of a single crystal layer.

2. The establishment of the developing procedures for manufacturing powder screens using new promising materials.

3. The selection of the scintillator materials, which are most suitable for each imaging application (i.e. exhibiting high light emission efficiency and optimized imaging performance).

4. The selection of the optimal crystal geometries for gamma ray imaging applications, which will provide a compromise between cost and performance.

5. The implementation of a well-tested simulation code, for studying optical phenomena in single and powder crystals.

In addition, important exchange of knowledge and experience from the collaboration of the proposing institutions is gained and subparts of this work lead to scientifically interesting results.

METHODS

The scintillator materials are commercially supplied n various physical forms (powders, single crystals and, if available, in ceramic form). In the case of powder scintillators, phosphor screens of various coating thicknesses will be prepared in the leader Institution by employing sedimentation techniques. Manufacturing of crystals surface or edges, is carried out by experienced personnel with associated infrastructure.

Evaluation of the performance of the scintillators is accomplished by determining the following parameters using experimental and theoretical methods:

1. Absolute luminescence efficiency (AE). This efficiency is defined as the ratio of the light energy flux emitted by an irradiated scintillator over the exposure rate characterizing the incident radiation. AE expresses the sensitivity of a scintillator and is of importance when the final image brightness with respect to the patient radiation dose is considered. The emitted light flux will be experimentally determined under x-ray and gamma-ray irradiation conditions. In addition, theoretical models, describing the radiation and light transmission through a scintillator material, will be employed to fit theoretical curves to experimental data. This technique will allow for estimation of a number of intrinsic material properties related to the light generation and the light attenuation within the mass of the scintillator.

2. Energy resolution (ER). This parameter expresses the ability of a scintillator to discern X-ray or gamma-ray photons of different energies. Detectors with good energy resolution produce high information diagnostic images and can be utilized in gamma-ray spectroscopy measurements

3. Optical spectrum (OS) emitted by an irradiated scintillator and the spectral compatibility (SC) of this spectrum with the spectral sensitivity of various optical sensors (photocathodes, photodiodes, films etc) used currently used in MI. SC is very important for estimating the efficiency of the detection of the emitted light of a new scintillator by existing optical sensors. A high value of SC ensures a lower dose to the patient and a faster Nuclear Medicine examination.

4. Modulation transfer function (MTF). This is an imaging parameter, which describes the image contrast and spatial resolution in the spatial frequency domain. Methods (SWRF response, ESF, etc) have been developed for measuring MTF of screen shaped scintillators (i.e. thin layers). In addition theoretical models, describing the radiation and light transmission through a scintillator material, will be employed to predict the MTF curves of the scintillator in various coating thicknesses and X-ray or gamma ray energies.

5. Noise power spectrum (NPS) or Wiener spectrum, which describes the noise contained ίn the final image. NPS can be calculated as the Fourier Transform of the autocorrelation function of the output signal of the scintillator detector under uniform X-ray or gamma-ray excitation. As in the case of MTF, theoretical models will be employed predict the NPS curves of the scintillator in various coating thicknesses and X-ray or gamma ray energies.

6. Detective quantum efficiency (DQE) describing the efficiency of an imaging system to transfer the input signal to noise ratio to its output. DQE values in the spatial frequency domain are a function of the MTF and the NPS of the scintillation detector. Since, DQE incorporates both signal and noise transfer it can be used as a single parameter for evaluating the performance of scintillators.

7. Measurement of the temporal response of certain scintillators and the effects of screen preparations upοn this process. Scintillators with a fast temporal response are of great value in the design of a ring type SPECT detector, especially in dynamic imaging, where the patient’s internal organs (i.e. heart) are moving.

8. Spatial resolution (SR). This parameter expresses the minimum distance that two X-ray or gamma-ray photons must have in order to be detected as two separate events by a scintillator. Detectors with good spatial resolution produce detailed images and can be utilized in gamma-ray imaging in order to identify small tumours.

The aforementioned parameters are determined under different exposure conditions used in variοus x-ray imaging techniques such as mammography, general radiography and fluoroscopy, computed tomography etc.

EXPERIENCE

The research group has a considerably long experience in investigating the role of powder (granular) and single crystal scintillators in the performance of radiation detectors of MI systems. More than 100 scientific papers, relevant to this field, have been published in international journals and conference proceedings. Various commercially available scintillator materials (terbium, europium and cerium activated rare earth materials as well as cesium iodide crystals) have been evaluated under x-ray and gamma-ray exposure conditions using experimental, theoretical as well as Monte Carlo techniques. Powder scintillators are employed in the form of thin screens prepared in laboratory with various thicknesses. Various properties related to optical characteristics (luminescence emission efficiency, light emission spectrum etc) and to the role of scintillators in the imaging performance (MTF, NPS, SNR, DQE etc) of imaging systems have been investigated. However the TEI group is not yet experienced in the development of imaging hardware and in incorporating scintillator materials in real detector systems. Taking into consideration the experience of ICCS in developing integrated imaging devices and the experience of the USA partner in producing and treatment novel scintillators, the present collaboration will clearly lead to benefits for all participating institutions by combining their complementary experiences.

DISSEMINATION AND EXPLOITATION OF RESEARCH RESULTS

The present research covers a variety of fields related to crystal evaluation, nano-technology, signal processing and MI. Since the techniques to be used are rather new and promising, a number of publications are expected to result from this collaboration. The results are related to the following: 1) Comparative evaluation of commercial used crystal materials in powder and single crystal form in x-ray and gamma-ray radiation, 2) evaluation of novel materials in powder and single crystal form, 3) Crystal treatment on micro-level, 4) Results from Monte Carlo studies and comparison with experimental ones.

REFERENCES

Antonuk L. E., Jee K. W., El-Mohri Y., Maolinbay M., Naddif S., Rong X., Zhao Q. and Siewerdsen J. H. “Strategies to improve the signal and noise performance of active matrix, flat-panel imagers for diagnostic x-ray applications,” Med. Phys. 27 (2000) 289-306.

Arvanitis C D, Bohndiek S E, Royle G, Blue A, Liang H X, Clark A, Prydderch M, Turchetta R, Speller R, “Empirical electro-optical and x-ray performance evaluation of CMOS active pixels sensor for low dose, high resolution x-ray medical imaging,” Med. Phys. 34 (2007) 4612-4625.

Badano A. and Sempau J., “MANTIS: combined x-ray, electron and optical Monte Carlo simulations of indirect radiation imaging systems” Phys. Med. Biol. 51(2006) 1545-1561.

Blasse G. and Grabmaier B. C., Luminescent materials (Spinger, Berlin, 1994).

Bohren C. F. and Huffman D. R., Absorption and Scattering of Light by Small Particles (Wiley, New York, 1983).

David S., Michail C., Valais I., Toutountzis A., Liaparinos P., Cavouras D., Kandarakis I., and Panayiotakis G., ‘Investigation of luminescence properties of Lu2SiO5:Ce (LSO) powder scintillator in the x-ray radiography energy range’, IEEE Transactions on Nuclear Science, 46, 474-478, 2007.

ICRU Report No. 41, Modulation Transfer Function of Screen-Film Systems, International Commission on Radiation Units and Measurements Inc., Bethesda, MD (1986).

International Electrotechnical Commission, International Standard IEC 62220-1:2003-10, "Medical Electrical Equipment-Characteristics of digital X-ray imaging devices-Part 1-2: Determination of the detective quantum efficiency-Mammography detectors, IEC, International Electrotechnical Commission, Geneva, Switzerland, 2005, IEC 62220-1-2.

Liaparinos P. and Kandarakis I., ‘The imaging performance of compact Lu₂O₃:Eu phosphor screens: Monte Carlo simulation for applications in mammography,’ Med.Phys., 36, 1985-1997, 2009,

Liaparinos P. and Kandarakis I., ‘The Monte Carlo evaluation of noise and resolution properties of granular phosphor screens,’ Phys. Med. Biol., 54, 859-874, 2009.

Michail C., David S., Liaparinos P., Valais I., Nikolopoulos D., Kalivas N., Toutountzis A., Sianoudis I., Cavouras D., Dimitropoulos N., Nomicos C. D., Kourkoutas K., Kandarakis I., Panayiotakis G. S., ‘Evaluation of the imaging performance of LSO powder scintillator for use in x-ray mammography’, Nucl. Instrum. Methods A, 580, 719-727, 2007.

Michail C, Spyropoulou V, Fountos G, Kalyvas N, I. G. Valais, I. S. Kandarakis and G. S. Panayiotakis,Experimental and theoretical evaluation of a high resolution CMOS based detector under X-ray imaging conditions, ΙΕΕΕTransactions on Nuclear Science (TNS), 58 (2011), 314-322.

Nagarkar V. V., Miller S. R., Tipnis S. V., Lempicki A., Brecher C., Lingertat H., “A new large area scintillator screen for X-ray imaging”, Nucl. Instr. and Meth. B. 213, (2004) 250-254.

Nikl M., “Scintillation detectors for x-rays,”Meas. Sci. Technol., vol. 17, pp.R37-R54, 2006.

Nikolopoulos D, Kandarakis I, CavourasD, Louizi A, Nomicos C. Investigation of radiation absorption and x-ray fluorescence of medical imaging scintillators by Monte Carlo Methods, Nucl.Instr.Method.(A) 2006; 565:821-832.

Valais I., Kandarakis I., Nikolopoulos D., Sianoudis I., Dimitropoulos N., Cavouras D., Nomicos C. D., and Panayiotakis G. S., “Luminescence efficiency of Gd2SiO5:Ce scintillator under x-ray excitation,” IEEE Trans. Nucl. Sci. 52 (2005) 1830-1835.

Van Eijk, “Inorganic scintillators in medical imaging,” Phys. Med. Biol. 47 (2002) R85-R106.

Van de Hulst H. C., Light Scattering by Small Particles (Wiley, New York, 1957).

Williams M. B., Simoni P. U., Smilowitz L., Stanton M., Phillips W. and Stewart A. Analysis of the detective quantum efficiency of a developmental detector for digital mammography Med. Phys. 26 (1999) 2273-2285.

Yaffe M. J. and Rowlands J. A, “X-ray detectors for digital radiography,” Phys. Med. Biol.42 (1997) 1-39.

Zych E, Meijerink A and Celso de Mello Doneg, “Quantum efficiency of europium emission from nanocrystalline powders of Lu₂O₃:Eu,” J. Phys.: Condens. Matter 15 (2003) 5145–5155.

Zych E, Trojan-Piegza J, HreniakD and Strek W, “Properties of Tb-doped vacuum-sintered Lu₂O₃storage phosphor nanocrystalline powders of Lu₂O₃:Eu,” J. Appl. Phys., 94 (2003) 1318–1324.