Sergey Komarov

A sub-millimeter resolution PET detector module using a multi-pixel photon counter array

The aim of this study is to design a novel PET detector module using an array of sub-millimeter LSO crystal read out by an array of surface-mount type of Multi-Pixel Photon Counter (MPPC). The volume of one MPPC sensor is 2.4 × 1.9 × 0.8 mm3 and the sensitive area is 1 × 1 mm2. The 9 MPPCs were arranged into a 3 by 3 array held by a Teflon base with 9 pockets. This provides an 8.6 × 8.6 mm2 sensing area to decode a 10 by 10 array with 0.8 × 0.8 × 3 mm3 LSO crystals of 0.86 mm pitch. After all the components were assembled, the base was mounted to a readout board that consists of quenching resistors, coupling capacitors, resistive charge divider, and current feedback pre-amplifiers. A light guide was fabricated to control the distribution of scintillation lights generated from the LSO array on the MPPC sensitive area so as to get a better localization of each gamma-ray event. A Monte Carlo simulation program was utilized to assist the light guide design. For the flood image measurement, a Na-22 point source was used. The energy resolution was measured with a Ge-68 source for each LSO crystal by a Gaussian fit to the photopeak in its energy spectrum. Timing resolution was measured against a plastic scintillator coupled to a Hamamatsu H5783 PMT using a standard time-to-amplitude converter. Linearity test was performed with gamma-ray sources of various energies. The 10 by 10 array of 0.8 mm LSO crystals can be clearly resolved in the flood image. Timing resolution of the single channel MPPC was estimated to be 620 picoseconds FWHM. Mean energy resolution and standard deviation value were 18.8% FWHM and ± 2.7% at 511 keV. The nonlinearity is observed at the single MPPC and the corner crystals in the MPPC array, but less significant for central crystal. These results demonstrate that the proposed PET detector module can be used for high resolution PET insert applications. It can potentially be used for MR compatible PET applications.

3-D Spatial Resolution of 350 /spl mu/m Pitch Pixelated CdZnTe Detectors for Imaging Applications

We are currently investigating the spatial resolution of highly pixelated Cadmium Zinc Telluride (CZT) detector for imaging applications. A 20 mm {\times} 20 mm {\times} 5 mm CZT substrate was fabricated with 600 {\mu}m pitch pixels (500 {\mu}m anode pixels with 100 {\mu}m gap) and coplanar cathode. Charge sharing between two pixels was studied using collimated 122 keV gamma ray source. Experiments show a resolution of 125 {\mu}m FWHM for double-pixel charge sharing events when the 600 {\mu}m pixelated and 5 mm thick CZT detector biased at -1000 V. In addition, we analyzed the energy response of the 600 {\mu}m pitch pixelated CZT detector.We are currently investigating the feasibility of using highly pixelated Cadmium Zinc Telluride (CdZnTe) detectors for sub-500 μm resolution PET imaging applications. A 20 mm × 20 mm × 5 mm CdZnTe substrate was fabricated with 350 μm pitch pixels (250 μm anode pixels with 100 μm gap) and coplanar cathode. Charge sharing among the pixels of a 350 μm pitch detector was studied using collimated 122 keV and 511 keV gamma ray sources. For a 350 μm pitch CdZnTe detector, scatter plots of the charge signal of two neighboring pixels clearly show more charge sharing when the collimated beam hits the gap between adjacent pixels. Using collimated Co-57 and Ge-68 sources, we measured the count profiles and estimated the intrinsic spatial resolution of 350 μm pitch detector biased at -1000 V. Depth of interaction was analyzed based on two methods, i.e., cathode/anode ratio and electron drift time, in both 122 keV and 511 keV measurements. For single-pixel photopeak events, a linear correlation between cathode/anode ratio and electron drift time was shown, which would be useful for estimating the DOI information and preserving image resolution in CdZnTe PET imaging applications.

A high resolution PET insert system for clinical PET/CT scanners

We have developed a PET insert system that can be integrated into a clinical PET-CT scanner to improve its image resolution when imaging a patient for head and neck cancer. The system consists of 28 detector modules arranged into two half rings (24.6 cm in diameter). Each detector consists of an array of 13×13 LSO crystals each measuring 2 × 2 × 5 mm3, a 2.8 mm thick light-guide, and a R8900-M16 PMT (Hamamatsu, Japan). The 16 anode outputs of a PMT are multiplexed to produce 4 position-encoded signals that resemble the output of a typical block detector in the PET scanner. An active switch board was designed to select signals from either the insert detectors or scanner detectors and feed the output signals to system electronics for event processing, allowing the scanner to switch between its normal operation and the insert mode without shutting down the system. A 3-D translation stage supports and positions the insert concentrically inside the scanner. The axial extend of the insert detectors is 54 mm, centered within the first 3 detector block rings of the scanner. Initial setup of the PET insert system requires the identification of crystal lookup table and photopeak locations, timing alignment of the insert detectors with the scanner. Once completed, the scanner can be switched between the insert and normal modes within a few minutes. Coincidences are sorted into insert-insert (II), insert-scanner (IS) and scanner-scanner (SS) types of events using a custom sorting code. We have developed 2D FBP and 2D ML-EM algorithms to reconstruct single-slice rebinned sinograms, and a 3D OS-EM algorithm to jointly estimate the images using all 3 types of events. Initial tests show a mean, max and min energy resolution of 12.36%, 18.57% and 9.45% FWHM at 511 keV, respectively. Image resolution measured by 22Na point sources is ∼2–3 mm FWHM and improved from 4–5mm FWHM for scanner system. Full system performance is currently being evaluated.

Image reconstruction and system modeling techniques for virtual-pinhole PET insert systems

Virtual-pinhole PET (VP-PET) imaging is a new technology in which one or more high-resolution detector modules are integrated into a conventional PET scanner with lower resolution detectors. It can locally enhance the spatial resolution and contrast recovery near the add-on detectors, and depending on the configuration, may also increase the sensitivity of the system. This novel scanner geometry makes the reconstruction problem more challenging compared to the reconstruction of data from a stand-alone PET scanner, as new techniques are needed to model and account for the non-standard acquisition. In this paper, we present a general framework for fully 3D modeling of an arbitrary VP-PET insert system. The model components are incorporated into a statistical reconstruction algorithm to estimate an image from the multi-resolution data. For validation, we apply the proposed model and reconstruction approach to one of our custom-built VP-PET systems-a half-ring insert device integrated into a clinical PET/CT scanner. Details regarding the most important implementation issues are provided. We show that the proposed data model is consistent with the measured data, and that our approach can lead to reconstructions with improved spatial resolution and lesion detectability.

Compton Scattering in Clinical PET/CT With High Resolution Half Ring PET Insert Device

The integration of a high resolution PET insert into a conventional PET system can significantly improve the resolution and the contrast of its images within a reduced imaging field of view. For the rest of the scanner imaging field of view, the insert is a highly attenuating and scattering media. In order to use all available coincidence events (including coincidences between 2 detectors in the original scanner, namely the scanner-scanner coincidences), appropriate scatter and attenuation corrections have to be implemented. In this work, we conducted a series of Monte Carlo simulations to estimate the composition of the scattering background and the importance of the scatter correction. We implemented and tested the Single Scatter Simulation (SSS) algorithm for a hypothetical system and show good agreement between the estimated scatter using SSS and Monte Carlo simulated scatter contribution. We further applied the SSS to estimate scatter contribution from an existing prototype PET insert for a clinical PET/CT scanner. The results demonstrated the applicability of SSS to estimate the scatter contribution within a clinical PET/CT system even when there is a high resolution half ring PET insert device in its imaging field of view.

Characterization of highly pixelated CZT detectors for sub-millimeter PET imaging

Cadmium Zinc Telluride (CZT) detectors with 600 μm pitches and 350 μm pitches in 5mm thickness were developed and tested for investigation pixilated CZT detector applications in Positron Emission Tomography. In addition to flood γ-ray source measurements (662keV, 122keV and 59.5keV), we scan across pixels with a collimated Co-57 source (collimated beam size < pixel size) with a step size 50 μm. We sorted events based on the number of pixels that detect a charge signal (single, double, triple) and analyzed the charge sharing at different energies for detectors of different pitches. Results show the number of charge sharing events increases when γ-ray energy increases or when pixel size decreases. Based on the collimated beam measurements of energy at 122keV, the charge sharing range in a 5mm thick CZT detector biased at -1000V was estimated to be around 125μm FWHM for both 600μm and 350μm pitched anodes, while the charge loss is more severe in 350μm pitched detectors than in 600μm pitched detectors.

Penetration depth of photons in biological tissues from hyperspectral imaging in shortwave infrared in transmission and reflection geometries

Abstract. Measurement of photon penetration in biological tissues is a central theme in optical imaging. A great number of endogenous tissue factors such as absorption, scattering, and anisotropy affect the path of photons in tissue, making it difficult to predict the penetration depth at different wavelengths. Traditional studies evaluating photon penetration at different wavelengths are focused on tissue spectroscopy that does not take into account the heterogeneity within the sample. This is especially critical in shortwave infrared where the individual vibration-based absorption properties of the tissue molecules are affected by nearby tissue components. We have explored the depth penetration in biological tissues from 900 to 1650 nm using Monte–Carlo simulation and a hyperspectral imaging system with Michelson spatial contrast as a metric of light penetration. Chromatic aberration-free hyperspectral images in transmission and reflection geometries were collected with a spectral resolution of 5.27 nm and a total acquisition time of 3 min. Relatively short recording time minimized artifacts from sample drying. Results from both transmission and reflection geometries consistently revealed that the highest spatial contrast in the wavelength range for deep tissue lies within 1300 to 1375 nm; however, in heavily pigmented tissue such as the liver, the range 1550 to 1600 nm is also prominent.

Investigation of the limitations of the highly pixilated CdZnTe detector for PET applications

We are investigating the feasibility of a high resolution positron emission tomography (PET) insert device based on the CdZnTe detector with 350 µm anode pixel pitch to be integrated into a conventional animal PET scanner to improve its image resolution. In this paper, we have used a simplified version of the multi pixel CdZnTe planar detector, 5 mm thick with 9 anode pixels only. This simplified 9 anode pixel structure makes it possible to carry out experiments without a complete application-specific integrated circuits readout system that is still under development. Special attention was paid to the double pixel (or charge sharing) detections. The following characteristics were obtained in experiment: energy resolution full-width-at-half-maximum (FWHM) is 7% for single pixel and 9% for double pixel photoelectric detections of 511 keV gammas; timing resolution (FWHM) from the anode signals is 30 ns for single pixel and 35 ns for double pixel detections (for photoelectric interactions only the corresponding values are 20 and 25 ns); position resolution is 350 µm in x,y-plane and ∼0.4 mm in depth-of-interaction. The experimental measurements were accompanied by Monte Carlo (MC) simulations to find a limitation imposed by spatial charge distribution. Results from MC simulations suggest the limitation of the intrinsic spatial resolution of the CdZnTe detector for 511 keV photoelectric interactions is 170 µm. The interpixel interpolation cannot recover the resolution beyond the limit mentioned above for photoelectric interactions. However, it is possible to achieve higher spatial resolution using interpolation for Compton scattered events. Energy and timing resolution of the proposed 350 µm anode pixel pitch detector is no better than 0.6% FWHM at 511 keV, and 2 ns FWHM, respectively. These MC results should be used as a guide to understand the performance limits of the pixelated CdZnTe detector due to the underlying detection processes, with the understanding of the inherent limitations of MC methods.

System modeling of a DOI-capable half-ring PET insert device for breast cancer imaging

We have developed a new type of PET insert system that, when integrated into a whole-body PET scanner, offers the possibility of imaging the breast and chest wall region with higher resolution and sensitivity than a whole-body PET scanner alone. The focus of this paper is to describe a system model that can be used within a fully 3D iterative reconstruction algorithm to accurately model the measured data. We have modeled the most important components of the system matrix, including the spatially-variant geometric factor, inter-crystal penetration, and body attenuation. Due to the irregular sampling of the insert system, we compute the system matrix using the native detector geometry. Based on subdividing each detector volume into subvolumes, we show that this approach can model DOI crystals and non-DOI crystals under a common framework. We present some of our first results using data acquired by the physical PET insert system. More detailed modeling or normalization strategies will need to be employed to obtain the most benefit from the reconstruction algorithm.

Simulation study of charge collection in highly pixilated CdZnTe detector for PET imaging

In this simulation study, we consider a planar CdZnTe detector with pixilated anode as an insert detector for the conventional PET scanner. The anode size is mostly limited by the charge sharing between neighboring pixels. This charge sharing is affected by the following: diffusion of charge particles; original charge particles distribution after the interaction; multiple interactions of gamma rays inside the detector; and electronic crosstalk. To evaluate charge sharing in highly pixilated CdZnTe detectors, we developed the Monte Carlo (MC) simulation tool to guide the experiment design and validate its results. The MC tool includes a gamma tracking module and a signal processing module. The charge distribution after the interaction of gamma with CdZnTe is simulated using EGSnrc code [4]. The simulations were done for CdZnTe detector with cathode-anode distance L=5 mm and anode pixel size W=600 μm.