Polad M. Shikhaliev

Energy-resolved computed tomography : first experimental results

First experimental results with energy-resolved computed tomography (CT) are reported. The contrast-to-noise ratio (CNR) in CT has been improved with x-ray energy weighting for the first time. Further, x-ray energy weighting improved the CNR in material decomposition CT when applied to CT projections prior to dual-energy subtraction. The existing CT systems use an energy (charge) integrating x-ray detector that provides a signal proportional to the energy of the x-ray photon. Thus, the x-ray photons with lower energies are scored less than those with higher energies. This underestimates contribution of lower energy photons that would provide higher contrast. The highest CNR can be achieved if the x-ray photons are scored by a factor that would increase as the x-ray energy decreases. This could be performed by detecting each x-ray photon separately and measuring its energy. The energy selective CT data could then be saved, and any weighting factor could be applied digitally to a detected x-ray photon. The CT system includes a photon counting detector with linear arrays of pixels made from cadmium zinc telluride (CZT) semiconductor. A cylindrical phantom with 10.2 cm diameter made from tissue-equivalent material was used for CT imaging. The phantom included contrast elements representing calcifications, iodine, adipose and glandular tissue. The x-ray tube voltage was 120 kVp. The energy selective CT data were acquired, and used to generate energy-weighted and material-selective CT images. The energy-weighted and material decomposition CT images were generated using a single CT scan at a fixed x-ray tube voltage. For material decomposition the x-ray spectrum was digitally spilt into low- and high-energy parts and dual-energy subtraction was applied. The x-ray energy weighting resulted in CNR improvement of calcifications and iodine by a factor of 1.40 and 1.63, respectively, as compared to conventional charge integrating CT. The x-ray energy weighting was also applied to low- and high-energy CT projections used for material decomposition. This improved the CNR in images of decomposed calcification and iodine by a factor of 1.57 and 1.46, respectively, as compared to conventional charge integrating CT. Some limitations were observed due to hole trapping in CZT and charge sharing between the detector pixels. First experimental results demonstrate that energy-resolved CT is coming close to its practical applications. Although hole trapping and charge sharing in CZT deteriorates x-ray spectrum and limits CNR improvement with energy weighting and detector count rate, this problem has a feasible solution, which is discussed in this paper and is a matter of ongoing research.

Computed tomography with energy-resolved detection: a feasibility study.

The feasibility of computed tomography (CT) with energy-resolved x-ray detection has been investigated. A breast CT design with multi slit multi slice (MSMS) data acquisition was used for this study. The MSMS CT includes linear arrays of photon counting detectors separated by gaps. This CT configuration allows for efficient scatter rejection and 3D data acquisition. The energy-resolved CT images were simulated using a digital breast phantom and the design parameters of the proposed MSMS CT. The phantom had 14 cm diameter and 50/50 adipose/glandular composition, and included carcinoma, adipose, blood, iodine and CaCO3 as contrast elements. The x-ray technique was 90 kVp tube voltage with 660 mR skin exposure. Photon counting, charge (energy) integrating and photon energy weighting CT images were generated. The contrast-to-noise (CNR) improvement with photon energy weighting was quantified. The dual energy subtracted images of CaCO3 and iodine were generated using a single CT scan at a fixed x-ray tube voltage. The x-ray spectrum was electronically split into low- and high-energy parts by a photon counting detector. The CNR of the energy weighting CT images of carcinoma, blood, adipose, iodine, and CaCO3 was higher by a factor of 1.16, 1.20, 1.21, 1.36 and 1.35, respectively, as compared to CT with a conventional charge (energy) integrating detector. Photon energy weighting was applied to CT projections prior to dual energy subtraction and reconstruction. Photon energy weighting improved the CNR in dual energy subtracted CT images of CaCO3 and iodine by a factor of 1.35 and 1.33, respectively. The combination of CNR improvements due to scatter rejection and energy weighting was in the range of 1.71-2 depending on the type of the contrast element. The tilted angle CZT detector was considered as the detector of choice. Experiments were performed to test the effect of the tilting angle on the energy spectrum. Using the CZT detector with 20 degrees tilting angle decreased the tailing of the measured x-ray spectrum as compared to a conventional CZT detector. It was concluded that the energy-resolved MSMS CT with tilted angle CZT detector is potentially feasible and could provide a unique combination of photon counting, energy weighting, scatter rejection and single kVp dual energy subtraction CT imaging.

Photon counting multienergy x‐ray imaging: Effect of the characteristic x rays on detector performance

PURPOSE The purpose of this work was to investigate the effect of characteristic x rays on the performance of photon counting detectors for multienergy x-ray imaging. X-ray and CT systems with photon counting detectors have compelling advantages compared to energy integrating detectors, and cadmium zinc telluride (CZT) detector is the detector of choice. However, current CZT detectors exhibit several limitations that hamper their practical applications. These limitations include hole trapping, high leakage current, and charge sharing between detector pixels. Charge sharing occurs due to the diffusion of charge when it drifts toward the pixel electrodes. It also occurs due to nonlocal reabsorption of characteristic and scattered x rays created in the detector volume. Hole trapping, leakage current, and charge diffusion may potentially have technical solutions. Characteristic x-ray escape and scatter, however, are fundamental in nature and cannot be easily addressed. The x-ray scatter in the CZT material is small at photon energies used in x-ray imaging. Therefore, the remaining major factor is characteristic x ray. METHODS Monte Carlo simulations were used for this study. An experimental photon counting multienergy x-ray imaging system was used to compare simulations to experimental results. An x-ray spectrum at 120 kVp tube voltage was used. The x-ray energy range was split into five subregions (energy bins) and Monte Carlo simulations were performed at average x-ray energies corresponding to these energy bins. The detector pixel size was changed within the 0.1-1 mm range, which covered all possible applications including radiography and CT imaging. The pixel shapes included square and strip pixels. For strip pixels, tilted angle irradiation of the CZT detector was also investigated. RESULTS The characteristic x rays escaped the pixels in approximately 70% of all x-ray interactions for the smallest pixel size of 0.1 mm. The escape fraction decreased to 20% for the largest pixel size of 1 mm. All escape fractions, for all pixel sizes, at five energies, for square and strip pixels, and at three tilt angles were calculated and presented in tables. Simulated and measured spectra at 120 kVp were compared. CONCLUSIONS Characteristic x-ray escape deteriorates energy and spatial resolution, particularly for small pixel sizes. Correction methods should be developed based on the results of the simulations and experimental study.

Projection x-ray imaging with photon energy weighting: experimental evaluation with a prototype detector

The signal-to-noise ratio (SNR) in x-ray imaging can be increased using a photon counting detector which could allow for rejecting electronics noise and for weighting x-ray photons according to their energies. This approach, however, was not feasible for a long time because photon counting x-ray detectors with very high count rates, good energy resolution and a large number of small pixels were required. These problems have been addressed with the advent of new detector materials, fast readout electronics and powerful computers. In this work, we report on the experimental evaluation of projection x-ray imaging with a photon counting cadmium-zinc-telluride (CZT) detector with energy resolving capabilities. The detector included two rows of pixels with 128 pixels per row with 0.9 x 0.9 mm(2) pixel size, and a 2 Mcount pixel(-1) s(-1) count rate. The x-ray tube operated at 120 kVp tube voltage with 2 mm Al-equivalent inherent filtration. The x-ray spectrum was split into five regions, and five independent x-ray images were acquired at a time. These five quasi-monochromatic x-ray images were used for x-ray energy weighting and material decomposition. A tissue-equivalent phantom was used including contrast elements simulating adipose, calcifications, iodine and air. X-ray energy weighting improved the SNR of calcifications and iodine by a factor of 1.32 and 1.36, respectively, as compared to charge integrating. Material decomposition was performed by dual energy subtraction. The low- and high-energy images were generated in the energy ranges of 25-60 keV and 60-120 keV, respectively, by combining five monochromatic image data into two. X-ray energy weighting was applied to low- and high-energy images prior to subtraction, and this improved the SNR of calcifications and iodine in dual energy subtracted images by a factor of 1.34 and 1.25, respectively, as compared to charge integrating. The detector energy resolution, spatial resolution, linearity, count rate, noise and image uniformity were investigated. The limitations of this technology were emphasized and possible solutions were discussed.

Photon counting spectral CT: improved material decomposition with K-edge-filtered x-rays

Photon counting spectral computed tomography (PCSCT) provides material selective CT imaging at a single CT scan and fixed tube voltage. The PCSCT data are acquired in several energy ranges (bins) arranged over the x-ray spectrum. The quasi-monoenergetic CT images are acquired in these energy bins and are used for material decomposition. The PCSCT exhibits inherent limitations when material decomposition is performed using energy bins. For effective material decomposition, the energy bins used for material decomposition should be sufficiently narrow and well separated. However, when narrow bins are used, a large fraction of the detected x-ray counts is lost and statistical noise is increased. Alternatively, the x-ray spectrum can be split into a few larger bins with no gap in between and all detected x-ray photons can be used for material decomposition. However, in this case the energy bins are too wide and not well separated, which results in suboptimal material decomposition. The above contradictory requirements can be resolved if the x-ray photons are physically removed from the regions of the energy spectrum between the energy bins. Such a selective removal can be performed using filtration of the x-ray beam by high-Z filter materials with appropriate positions of K-edge energies. The K-edge filtration of x-rays can, therefore, provide necessary gaps between the energy bins with no dose penalty to the patient. In the current work, we proposed using selective K-edge filtration of x-rays in PCSCT and performed the first experimental investigation of this approach. The PCSCT system included a cadmium zinc telluride semiconductor detector with 2 × 256 pixels and 1 × 1 mm(2) pixel size, and five energy bins. The CT phantom had 14 cm diameter and included contrast elements of iodine, gold and calcifications with clinically relevant concentrations. The tube voltages of 60, 90 and 120 kVp were used. K-edge filters based on Ba (E(k) = 37.44 keV) were used for a 60 kVp tube voltage and Gd (E(k) = 50.24 keV) was used for the 90 and 120 kVp tube voltages, respectively. The material selective CT images were also acquired with conventional Al filtration for comparison. The half-value layers of x-ray beams after K-edge and Al filtration were matched. The mean entrance skin exposure was 280 mR for all tube voltages and filters. The contrast-to-noise ratio (CNR) in material-decomposed images was approximately 30%-50% higher when K-edge filters were used instead of Al filters. It was concluded that K-edge filtration of x-rays provides substantial improvement of the CNR in material-selective PCSCT. Further optimization of K-edge filter materials, tube voltages, detector technology and energy bin settings will provide even higher CNR in decomposed images.

The upper limits of the SNR in radiography and CT with polyenergetic x-rays

The aim of the study is to determine the upper limits of the signal-to-noise ratio (SNR) in radiography and computed tomography (CT) with polyenergetic x-ray sources. In x-ray imaging, monoenergetic x-rays provide a higher SNR compared to polyenergetic x-rays. However, the SNR in polyenergetic x-ray imaging can be increased when a photon-counting detector is used and x-rays are optimally weighted according to their energies. For a particular contrast/background combination and at a fixed x-ray entrance skin exposure, the SNR in energy-weighting x-ray imaging depends on tube voltage and can be maximized by selecting the optimal tube voltage. The SNR in energy-weighted x-ray images acquired at this optimal tube voltage is the highest SNR that can be achieved with polyenergetic x-ray sources. The optimal tube voltages and the highest SNR were calculated and compared to the SNR of monoenergetic x-ray imaging. Monoenergetic, energy-weighting polyenergetic and energy-integrating polyenergetic x-ray imagings were simulated at a fixed entrance skin exposure of 20 mR. The tube voltages varied in the range of 30-140 kVp with 10 kV steps. Contrast elements of CaCO(3), iodine, adipose and tumor with thicknesses of 280 mg cm(-2), 15 mg cm(-2), 1 g cm(-2) and 1 g cm(-2), respectively, inserted in a soft tissue background with 10 cm and 20 cm thicknesses, were used. The energy weighting also improves the contrast-to-noise ratio (CNR) in CT when monoenergetic CT projections are optimally weighted prior to CT reconstruction (projection-based weighting). Alternatively, monoenergetic CT images are reconstructed, optimally weighted and composed to yield a final CT image (image-based weighting). Both projection-based and image-based weighting methods improve the CNR in CT. An analytical approach was used to determine which of these two weighting methods provides the upper limit of the CNR in CT. The energy-weighting method was generalized and expanded as a weighting method applicable in areas other than x-ray and CT. An optimal x-ray tube voltage providing the highest SNR in energy-weighting x-ray imaging was determined for each contrast/background combination. Depending on the imaging task, the highest SNR with energy-weighted polyenergetic x-rays was close to the SNR provided by monoenergetic x-rays. In CT, projection-based weighting provided higher CNR than image-based weighting, thus determining an upper limit of the CNR in CT. The weighting approach can be applied to imaging methods with contrast mechanisms other than x-ray interaction. A unique, task-dependent tube voltage exists in photon-counting x-ray and CT that provides the highest SNR with polyenergetic x-rays. The highest SNR achieved in photon-counting energy-weighted x-ray and CT can approach the SNR of monoenergetic x-ray and CT imaging, depending on the imaging task.

Photon counting spectral breast CT: effect of adaptive filtration on CT numbers, noise, and contrast to noise ratio.

PURPOSE Photon counting spectral (PCS) computed tomography (CT) shows promise for breast imaging. An issue with current photon-counting detectors is low count rate capabilities, artifacts resulting from nonuniform count rate across the field of view, and suboptimal spectral information. These issues are addressed in part by using tissue-equivalent adaptive filtration of the x-ray beam. The purpose of the study was to investigate the effect of adaptive filtration on different aspects of PCS breast CT. METHODS The theoretical formulation for the filter shape was derived for different filter materials and evaluated by simulation and an experimental prototype of the filter was fabricated from a tissue-like material (acrylic). The PCS CT images of a glandular breast phantom with adipose and iodine contrast elements were simulated at 40, 60, 90, and 120 kVp tube voltages, with and without adaptive filter. The CT numbers, CT noise, and contrast-to-noise ratio (CNR) were compared for spectral CT images acquired with and without adaptive filters. Similar comparison was made for material-decomposed PCS CT images. RESULTS The adaptive filter improved the uniformity of CT numbers, CT noise, and CNR in both ordinary and material decomposed PCS CT images. At the same tube output the average CT noise with adaptive filter, although uniform, was higher than the average noise without adaptive filter due to x-ray absorption by the filter. Increasing tube output, so that average skin exposure with the adaptive filter was same as without filter, made the noise with adaptive filter comparable to or lower than that without adaptive filter. Similar effects were observed when energy weighting was applied, and when material decompositions were performed using energy selective CT data. CONCLUSIONS An adaptive filter decreases count rate requirements to the photon counting detectors which enables PCS breast CT based on commercially available detector technologies. Adaptive filter also improves image quality in PCS breast CT by decreasing beam hardening artifacts and by eliminating spatial nonuniformities of CT numbers, noise, and CNR.

CZT detectors used in different irradiation geometries: simulations and experimental results.

The purpose of this work was to evaluate potential advantages and limitations of CZT detectors used in surface-on, edge-on, and tilted angle irradiation geometries. Simulations and experimental investigations of the energy spectrum measured by a CZT detector have been performed using different irradiation geometries of the CZT. Experiments were performed using a CZT detector with 10 x 10 mm2 size and 3 mm thickness. The detector was irradiated with collimated photon beams from Am-241 (59.5 keV) and Co-57 (122 keV). The edge-scan method was used to measure the detector response function in edge-on illumination mode. The tilted angle mode was investigated with the radiation beam directed to the detector surface at angles of 90 degrees, 15 degrees, and 10 degrees. The Hecht formalism was used to simulate theoretical energy spectra. The parameters used for simulations were matched to experiment to compare experimental and theoretical results. The tilted angle CZT detector suppressed the tailing of the spectrum and provided an increase in peak-to-total ratio from 38% at 90 degrees to 83% at 10 degrees tilt angle for 122 keV radiation. The corresponding increase for 59 keV radiation was from 60% at 90 degrees to 85% at 10 degrees tilt angle. The edge-on CZT detector provided high energy resolution when the beam thickness was much smaller than the thickness of CZT. The FWHM resolution in edge-on illumination mode was 4.2% for 122 keV beam with 0.3 mm thickness, and rapidly deteriorated when the thickness of the beam was increased. The energy resolution of surface-on geometry suffered from strong tailing effect at photon energies higher than 60 keV. It is concluded that tilted angle CZT provides high energy resolution but it is limited to a 1D linear array configuration. The surface-on CZT provides 2D pixel arrays but suffers from tailing effect and charge build up. The edge-on CZT is considered suboptimal as it requires small beam thickness and also suffers from charge buildup.

Permanent-magnet energy spectrometer for electron beams from radiotherapy accelerators

PURPOSE The purpose of this work was to adapt a lightweight, permanent magnet electron energy spectrometer for the measurement of energy spectra of therapeutic electron beams. METHODS An irradiation geometry and measurement technique were developed for an approximately 0.54-T, permanent dipole magnet spectrometer to produce suitable latent images on computed radiography (CR) phosphor strips. Dual-pinhole electron collimators created a 0.318-cm diameter, approximately parallel beam incident on the spectrometer and an appropriate dose rate at the image plane (CR strip location). X-ray background in the latent image, reduced by a 7.62-cm thick lead block between the pinhole collimators, was removed using a fitting technique. Theoretical energy-dependent detector response functions (DRFs) were used in an iterative technique to transform CR strip net mean dose profiles into energy spectra on central axis at the entrance to the spectrometer. These spectra were transformed to spectra at 95-cm source to collimator distance (SCD) by correcting for the energy dependence of electron scatter. The spectrometer was calibrated by comparing peak mean positions in the net mean dose profiles, initially to peak mean energies determined from the practical range of central-axis percent depth-dose (%DD) curves, and then to peak mean energies that accounted for how the collimation modified the energy spectra (recalibration). The utility of the spectrometer was demonstrated by measuring the energy spectra for the seven electron beams (7-20 MeV) of an Elekta Infinity radiotherapy accelerator. RESULTS Plots of DRF illustrated their dependence on energy and position in the imaging plane. Approximately 15 iterations solved for the energy spectra at the spectrometer entrance from the measured net mean dose profiles. Transforming those spectra into ones at 95-cm SCD increased the low energy tail of the spectra, while correspondingly decreasing the peaks and shifting them to slightly lower energies. Energy calibration plots of peak mean energy versus peak mean position of the net mean dose profiles for each of the seven electron beams followed the shape predicted by the Lorentz force law for a uniform z-component of the magnetic field, validating its being modeled as uniform (0.542 ± 0.027 T). Measured Elekta energy spectra and their peak mean energies correlated with the 0.5-cm (7-13 MeV) and the 1.0-cm (13-20 MeV) R90 spacings of the %DD curves. The full-width-half-maximum of the energy spectra decreased with decreasing peak mean energy with the exception of the 9-MeV beam, which was anomalously wide. Similarly, R80-20 decreased linearly with peak mean energy with the exception of the 9 MeV beam. Both were attributed to suboptimal tuning of the high power phase shifter for the recycled radiofrequency power reentering the traveling wave accelerator. CONCLUSIONS The apparatus and analysis techniques of the authors demonstrated that an inexpensive, lightweight, permanent magnet electron energy spectrometer can be used for measuring the electron energy distributions of therapeutic electron beams (6-20 MeV). The primary goal of future work is to develop a real-time spectrometer by incorporating a real-time imager, which has potential applications such as beam matching, ongoing beam tune maintenance, and measuring spectra for input into Monte Carlo beam calculations.

TU‐E‐217BCD‐04: Spectral Breast CT: Effect of Adaptive Filtration on CT Numbers, CT Noise, and CNR

PURPOSE Photon counting spectral breast CT is feasible in part due to using an adaptive filter. An adaptive filter provides flat x-ray intensity profile and constant x-ray energy spectrum across detector surface, decreases required detector count rate, and eliminates beam hardening artifacts. However, the altered x-ray exposure profiles at the breast and detector surface may influence the distribution of CT noise, CT numbers, and contrast to noise ratio (CNR) across the CT images. The purpose of this work was to investigate these effects. METHODS Images of a CT phantom with and without adaptive filter were simulated at 60kVp, 90kVp, and 120kVp tube voltages and 660 mR total skin exposure. The CT phantom with water content had 14cm diameter, contrast elements representing adipose tissue and 2.5mg/cc iodine contrast located at 1cm, 3.5cm, and 6cm from center of the phantom. The CT numbers, CT noise, and CNR were measured at multiple locations for several filter/exposure combinations: (1)without adaptive filter for 660mR skin exposure; (2)with adaptive filter for 660mR skin exposure along central axis (mean skin exposure across the breast was <660mR); and (3)with adaptive filter for scaled exposure (mean skin exposure was 660mR). RESULTS Beam hardening (cupping) artifacts had 47HU magnitude without adaptive filter but were eliminated with adaptive filter. CNR of contrast elements was comparable for (1) and (2) over central parts but was higher by 20-30% for (1) near the edge of the phantom. CNR was higher by 20-30% in (3) as compared to (2) over central parts and comparable near the edges. CONCLUSIONS The adaptive filter provided: uniform distribution of CT noise, CNR, and CT numbers across CT images; comparable or better CNR with no dose penalty to the breast; and eliminated beam hardening artifacts.