M. Betti

Design of compact pinhole SPECT system based on flat panel PMT

The present development of new gamma imagers has allowed to realize detectors with ultra high spatial resolution and very compact size for PET as well as for SPET application. In this paper we analyze and discuss the possible design of new pinhole SPECT scanners based on heads which consist of flat panel PSPMT and different design of scintillation arrays like NaI(Tl), 1 mm pixel size, and CsI(Tl) multi layers array, mounted in off centered configuration to improve the intrinsic spatial resolution of the imagers. The results show that an array configuration 2/spl times/2 Hamamatsu flat panel PSPMTs coupled to NaI(Tl) scintillation array with 1 mm pixel size, represents the best trade off between compactness and spatial resolution of pinhole SPET scanner. The use of off centered CsI(Tl) scintillation array coupled to a single flat panel PSPMT allows to arrange a high sensitivity and very compact pinhole SPET scanner at very low cost only worsening of 50% spatial resolution than an Anger gamma camera pinhole SPECT.

Custom breast phantom for an accurate tumor SNR analysis

The capability of the scintimammography to diagnose subcentimeters sized tumors was increased by the employment of a dedicated gamma camera. The introduction of small field of view camera, based on pixellated scintillation array and position sensitive photomultiplier, allowed to enhance the geometric spatial resolution and contrast of the images due to reduced collimator-tumor distance. The aim of this paper is to investigate the realistic possibility of T1a tumors detection (/spl sim/5 mm size) by comparing the signal-to-noise ratio (SNR) values obtained by different imagers. To this end, we have utilized a self-designed solid breast phantom with different sized hot spots (tumors). The phantom consists of seven disks with different thickness, molded from resin epoxy activated with Co/sup 57/ isotope. The overlapped disks represent a pendula breast with about 800 cc volume. Hot spots have not wall. One disk has holes to fit the hot spots representing the different sized lesions. The imagers utilized were: a standard Anger Camera and three different cameras based on scintillator array, CsI(Tl) or NaI(Tl), coupled to position sensitive photomultiplier with different technologies, to make detectors with field of view of 3 and 5 inch. The experimental results are supported by Monte Carlo simulation. It was highlighted how spatial resolution is a predominant element in tumor visibility and how background causes a reduction of the image contrast. All gamma cameras show close results at SNR values less than 10 and a full detectability of 8 mm tumor size. However, the results show the 5 mm tumor size is lower detection limit for all cameras.

New devices for imaging in nuclear medicine.

Pinhole gamma camera imaging offers the ability to obtain high resolution images from single gamma ray emitting radiotracers playing a reasonable tradeoff between very small field of view (FoV) and sensitivity. On the other hand the total spatial resolution of a pinhole imaging device is predominantly affected by the detector intrinsic spatial resolution for reduced magnification factors. To design very compact pinhole SPET scanners with very high intrinsic spatial resolution, authors investigated a miniature gamma camera based on the newly developed Hamamatsu H8500 flat panel photomultiplier. The PSPMT was coupled to the following scintillation arrays: CsI(Tl) array with 0.2-mm, 1-mm, 1.4-mm pixel size and NaI (Tl) with 1-mm pixel size. The imaging performances were evaluated by 57Co spot and flood irradiations. NaI(Tl) array shows a better pixel identification for 1 mm pixel size, proving to be a good candidate to make a large area photodetector based on multi PSPMTs closely packed. Although CsI(Tl) array had the smallest pixel size, the low light output limited the best intrinsic spatial resolution to about 0.5 mm.

Imaging characteristics comparison of compact pixellated detectors for scintimammography

Although scintimammography is routinely used in the hospitals it had not the expected large diffusion. The low specificity of SESTAMIBI /sup 99m/Tc and the geometrical restrictions of prone scintimammography further contributed in limiting the clinical sensitivity for sub centimeter lesions. Researchers aware that to overcome intrinsic limitation of radio-pharmaceutical, they are optimized compact imager able to detect small amount of radioactivity emitted from 5 mm or less tumors. To investigate the overall imaging characteristics of scintillation imagers for scintimammography, a number of small gamma camera prototypes were assembled, based on PSPMT of first and last generation. They were coupled to a number of NaI(Tl) and CsI(Tl) scintillation both planar and pixellated, with different pixel size. Measurements were performed by line and circular source as well as by 3 mm to 10 mm diameter tumor phantoms. Spatial resolution seems the dominant factor, in detection of 5 mm and 3 mm tumor size. The best results were obtained from the scintillation crystals arrays coupled to Hamamatsu flat panel PMT. Decreasing crystal pixel size, contrast mean values slightly changed but their accuracy increases, demonstrating the influence of digitization on image quality. MTF curve obtained by planar NaI(Tl) integral assembly shows an improvement with a response at the highest frequency equal to one obtained from collimator parallel hole Monte Carlo simulation.

Imaging performances of LaCl/sub 3/:Ce scintillation crystals in SPECT

Over the last three years, there is a growing interest in the development of a new class of fast scintillators like LaCl/sub 3/:Ce and LaBr/sub 3/:Ce. The new scintillation crystals for the first time, have the main characteristic in producing a light photon number higher than NaI(Tl) at a wavelength suited for the photocathode. The most important characteristic of these crystals is represented from the scintillation light to be yield proportionality as a function of incident gamma ray energy. With the aim to investigate on their potential in single photon imaging three one inch square LaCl/sub 3/:Ce continuous scintillation crystals with 3 and 6 mm thickness were specifically designed for position measurements. They have been coupled to a flat panel PMT through a 3 mm thickness glass window. In order to highlight all the characteristics of the detection system the third one, 3 mm thick, was integral assembled with a one inch PSPMT Hamamatsu R5900-C8. Free and collimated point radioactive sources were used for energy resolution measurements as well for scanning crystals to investigate spatial resolution and position response. Energy resolution values were compared with ones obtained from a sample of LaBr/sub 3/:Ce with one inch diameter and one thickness designed for spectrometric measurements. LaCl/sub 3/:Ce with 3 mm thickness shown energy resolution values worse than expected (13% at 140 KeV) due to the sub-optimal size and configuration for light collection. On the contrary LaBr/sub 3/:Ce carried out 6% relative energy resolution at 140 keV. Intrinsic spatial resolution values of 1 mm and 1.3 mm were obtained from 3 mm thickness LaCl/sub 3/:Ce crystals in the two configurations investigated.

Results of clinical trials with SPEM

Abstract X-ray mammography represents the principal tool for breast cancer screening, but has several limitations. Due to the low specificity of X-ray mammography, many more biopsies are performed than are necessary. More than 60% of breast biopsies performed because of suspicious X-ray mammograms yield a diagnosis other than cancer. In women with X-ray dense breast tissue, X-ray mammography can miss as many as 20% of cancers. Several studies suggest that combining 99mTc Sestamibi Scintimammography (SM) with X-ray mammography can increase the accuracy of breast imaging in selected populations. The principal limitation of prone-position scintimammography (PSM) using a standard Anger gamma camera is low sensitivity for subcentimeter cancers (i.e., stages T1a and T1b). Detecting these small cancers is extremely important clinically, since removal of the cancers at these early stages is thought to represent the best opportunity for cure. Our group constructed a high spatial resolution detector specifically dedicated to SM, the Single Photon Emission Mammography (SPEM) camera. Unlike conventional Anger gamma cameras, the SPEM camera incorporates a high spatial resolution position-sensitive photomultiplier tube, coupled to an array of scintillating crystals. The compactness of the SPEM camera allows breast compression to be implemented in a cranio-caudal view, facilitating comparison to X-ray mammograms taken in the same position. Clinical results so obtained have demonstrated increased diagnostic sensitivity in sub-centimeter tumors (80% for SPEM vs 50% with PSM). Factors contributing to this increased sensitivity include improved signal-to-noise ratio (SNR). For sub-centimeter cancers, SPEM SNR values were consistently much higher than those of PSM. Classic detectability studies have demonstrated that an SNR value >5 is required for reliable detection of cancers. For subcentimeter cancers, only the SPEM attained or exceeded this minimum threshold. The results showed that compression, which optimizes resolution by reducing lesion-to-collimator distance, and increases planar image contrast by reducing signal and attenuation from interposed normal breast tissue, plays a fundamental role in the visualization of subcentimeter breast cancers.

Iodine 125 Imaging in Mice Using NaI(Tl)/Flat Panel PMT Integral Assembly

Radiolabeled agents that bind to specific receptors have shown great promise in diagnosing and characterizing tumor cell biology. A second area of interest is in vivo imaging of gene transcription and protein expression. The radioisotope I/sup 125/ is commercially available as a label for molecular probes and utilized by researchers in small animal studies. We propose an advanced imaging detector based on planar NaI(TI) integral assembly with a Hamamatsu Flat Panel PMT, representing one of the best trade-off between spatial resolution and detection efficiency. We tested the in-vivo performance of the detector by acquiring images of mice as a part of a study of inflammatory bowel disease (IBD). In this study, four 25 g mice with an IBD-like phenotype (SAMPl/YitFc) were injected with 375, 125, 60 and 30 /spl mu/Ci of /sup 125/I-labelled antibody against MAdCAM-1, a gut-specific endothelial cell adhesion molecule that is up-regulated in the presence of inflammation. Two mice without bowel inflammation were injected with 150 and 60 /spl mu/Ci of the labelled anti-MAdCAM-1 antibody as controls. To better evaluate the performances of the integral assembly detector, we also acquired mice images with a dual modality (X and Gamma Ray) camera dedicated for small animal imaging. The results of this new detector are impressive: images of SAMP1/YitFc injected with 30 /spl mu/Ci activity show inflammation throughout the intestinal tract, with the disease very well defined at two hours post-injection.

DISIS—Discrete Scintillation Imager Simulator

To foresee the response of a discrete scintillation imager, operating in single-photon modality, a computer simulation code has been developed and tested. The simulator, whose acronym is DISIS, may be used for designing and/or evaluating devices based on scintillation arrays coupled to a position sensitive light sensors by light guides. The simulator can be classified as a deterministic code because it is based on: (i) an analytical model describing the scintillation light distribution for charge integrals calculation, (ii) an algorithm using charge values for event centroiding and (iii) a Gaussian model for spreading the events on the image according to the inherent imager resolution. Particularly, DISIS allows one to study the influence of each setup parameter on the device spatial response. The imager optimization can be achieved by improving the pixel identification. To this aim the best trade-off has to be found between: (a) the light distribution at photocathode, (b) the light sampling capability over the photocathode area and (c) the centroiding algorithm

LaBr/sub 3/:Ce scintillation camera

Over the last three years, there has been a growing interest in the development of a new class of fast scintillators such as LaCl3 :Ce and LaBr3:Ce, or their superior energy resolution is opening an easier way to improve spatial resolution. In this paper we present the results obtained from the first LaBr3:Ce small gamma camera. It is based on continuous 50 times 50 mm2 crystal 5 mm thick integral assembled with a Hamamatsu Flat panel PMT. This detector optical configuration permits the narrowest light distribution and highest light collection to obtain the best spatial and energy resolution values respectively. At the same time 5 mm crystal thickness carries out 80% efficiency at 140 keV photon energy. Measurements of spatial resolution are also compared with the analogous ones obtained from another 50 times 50 mm2 LaBr3:Ce crystal, assembled with a 3 mm glass optical window. Energy resolution values have been furtherly compared with ones obtained from a LaBr3:Ce one inch diameter and thickness optimized for spectrometric measurements. The first LaBr3:Ce gamma camera shows excellent intrinsic spatial resolution values such as 0.9 mm with a best energy resolution value of 6.0% at 140 keV photon energy

Dependence of apparent diffusion coefficient measurement on diffusion gradient direction and spatial position – A quality assurance intercomparison study of forty-four scanners for quantitative diffusion-weighted imaging

PURPOSE To propose an MRI quality assurance procedure that can be used for routine controls and multi-centre comparison of different MR-scanners for quantitative diffusion-weighted imaging (DWI). MATERIALS AND METHODS 44 MR-scanners with different field strengths (1 T, 1.5 T and 3 T) were included in the study. DWI acquisitions (b-value range 0-1000 s/mm2), with three different orthogonal diffusion gradient directions, were performed for each MR-scanner. All DWI acquisitions were performed by using a standard spherical plastic doped water phantom. Phantom solution ADC value and its dependence with temperature was measured using a DOSY sequence on a 600 MHz NMR spectrometer. Apparent diffusion coefficient (ADC) along each diffusion gradient direction and mean ADC were estimated, both at magnet isocentre and in six different position 50 mm away from isocentre, along positive and negative AP, RL and HF directions. RESULTS A good agreement was found between the nominal and measured mean ADC at isocentre: more than 90% of mean ADC measurements were within 5% from the nominal value, and the highest deviation was 11.3%. Away from isocentre, the effect of the diffusion gradient direction on ADC estimation was larger than 5% in 47% of included scanners and a spatial non uniformity larger than 5% was reported in 13% of centres. CONCLUSION ADC accuracy and spatial uniformity can vary appreciably depending on MR scanner model, sequence implementation (i.e. gradient diffusion direction) and hardware characteristics. The DWI quality assurance protocol proposed in this study can be employed in order to assess the accuracy and spatial uniformity of estimated ADC values, in single- as well as multi-centre studies.