Task-based transferable deep-learning scatter correction in cone beam computed tomography: a simulation study.

Purpose: X-ray scatter significantly affects the image quality of cone beam computed tomography (CBCT). Although convolutional neural networks (CNNs) have shown promise in correcting x-ray scatter, their effectiveness is hindered by two main challenges: the necessity for extensive datasets and the uncertainty regarding model generalizability. This study introduces a task-based paradigm to overcome these obstacles, enhancing the application of CNNs in scatter correction. Approach: Using a CNN with U-net architecture, the proposed methodology employs a two-stage training process for scatter correction in CBCT scans. Initially, the CNN is pre-trained on approximately 4000 image pairs from geometric phantom projections, then fine-tuned using transfer learning (TL) on 250 image pairs of anthropomorphic projections, enabling task-specific adaptations with minimal data. 2D scatter ratio (SR) maps from projection data were considered as CNN targets, and such maps were used to perform the scatter prediction. The fine-tuning process for specific imaging tasks, like head and neck imaging, involved simulating scans of an anthropomorphic phantom and pre-processing the data for CNN retraining. Results: For the pre-training stage, it was observed that SR predictions were quite accurate (SSIM≥0.9). The accuracy of SR predictions was further improved after TL, with a relatively short retraining time (≈70 times faster than pre-training) and using considerably fewer samples compared to the pre-training dataset (≈12 times smaller). Conclusions: A fast and low-cost methodology to generate task-specific CNN for scatter correction in CBCT was developed. CNN models trained with the proposed methodology were successful to correct x-ray scatter in anthropomorphic structures, unknown to the network, for simulated data.

Crystal scatter effects in a large-area dual-panel Positron Emission Mammography system.

Positron Emission Mammography (PEM) is a valuable molecular imaging technique for breast studies using pharmaceuticals labeled with positron emitters and dual-panel detectors. PEM scanners normally use large scintillation crystals coupled to sensitive photodetectors. Multiple interactions of the 511 keV annihilation photons in the crystals can result in event mispositioning leading to a negative impact in radiopharmaceutical uptake quantification. In this work, we report the study of crystal scatter effects of a large-area dual-panel PEM system designed with either monolithic or pixelated lutetium yttrium orthosilicate (LYSO) crystals using the Monte Carlo simulation platform GATE. The results show that only a relatively small fraction of coincidences (~20%) arise from events where both coincidence photons undergo single interactions (mostly through photoelectric absorption) in the crystals. Most of the coincidences are events where at least one of the annihilation photons undergoes a chain of Compton scatterings: approximately 79% end up in photoelectric absorption while the rest (<1%) escape the detector. Mean positioning errors, calculated as the distance between first hit and energy weighted (assigned) positions of interaction, were 1.70 mm and 1.92 mm for the monolithic and pixelated crystals, respectively. Reconstructed spatial resolution quantification with a miniDerenzo phantom and a list mode iterative reconstruction algorithm shows that, for both crystal types, 2 mm diameter hot rods were resolved, indicating a relatively small effect in spatial resolution. A drastic reduction in peak-to-valley ratios for the same hot-rod diameters was observed, up to a factor of 14 for the monolithic crystals and 7.5 for the pixelated ones.

Experimental validation of the ANTS2 code for modelling optical photon transport in monolithic LYSO crystals.

In this work the scintillation energy spectra originating from the background radioactivity from polished monolithic lutetium yttrium oxyorthosilicate coupled to position-sensitive silicon photomultipliers (SiPM) was studied using the open source Monte Carlo simulation package ANTS2. Two crystal sizes, fully and partially covering the photosensor area, three surface crystal wrappings (black, specular or diffuse) and the full signal formation process in the photosensor were considered. The simulation results were validated with experimental data acquired under the same geometric and detector operating conditions. In all cases ANTS2 simulated spectra have very good agreement with experimental results, reproducing the expected shape, with correct onset and end at 88 and 1190 keV, respectively, as well as sharp edges at the reference energies of 88, 88 + 202, 88 + 307 and 88 + 202 + 307 keV. The normalized root-mean square error between simulated and measured spectra varied between 4.3% and 10.4%.

A dedicated phantom design for positron emission mammography performance evaluation.

A standard protocol for performance evaluation of positron emission mammography (PEM) systems has not yet been established. In this work we propose a methodology based on the design of specific phantoms for this imaging modality with component dimensions in accordance with typical breast lesion sizes together with the adaptation of current international protocols designed for clinical and preclinical positron emission tomographs (PET) systems. This methodology was used to evaluate the performance of the Flex Solo II PEM scanner in terms of spatial resolution, uniformity and contrast lesion detectability, recovery coefficients and spill-over ratios. Positron range effects were studied with 18F and 68Ga, which have very different energy spectra. Our results indicate that in-plane spatial resolution of the system is around 3.0 mm and 4.4 mm for 18F and 68Ga, respectively. Lesion detectability tests with sphere diameters between 4 and 10 mm confirmed that the PEM system can resolve all the spheres (hot or cold). Percent contrast values for 18F lie between 6%-38% and 34%-51% for hot- and cold- spheres, respectively; the corresponding intervals for 68Ga are lower, 4%-25% and 32%-44%. Regarding uniformity quantification, the system shows percentage standard deviations within 4.9%-5.7%, while the percent background variability measurements ranged between 6.7% and 10.9% for both radionuclides. Recovery coefficients measured with hot rod diameters between 1.5 and 9 mm, have values between 0.2-1.05 and 0.17-0.69 for 18F and 68Ga, respectively. Spill-over ratios have large values (0.22 in average) for both radionuclides. Our results indicate that the phantoms and the methodology developed in this work can serve as the basis for establishing an image quality protocol for the systematic evaluation of PEM systems, with a potential extension for performance evaluation of dedicated breastPET scanners.

Coincidence energy spectra due to the intrinsic radioactivity of LYSO scintillation crystals.

BACKGROUND: Lutetium oxyorthosilicate or lutetium yttrium oxyorthosilicate (LYSO) scintillation crystals used in most current PET scanner detectors contain 176Lu, which decays by beta emission to excited states of 176Hf accompanied by the emission of prompt gamma rays or internal conversion electrons. This intrinsic radioactivity can be self-detected in singles mode as a constant background signal that has an energy spectrum whose structure has been explained previously. In this work, we studied the energy spectrum due to the intrinsic radioactivity of LYSO scintillation crystals of two opposing detectors working in coincidence mode. The investigation included experimental data, Monte Carlo simulations and an analytical model. RESULTS: The structure of the energy spectrum was completely understood and is the result of the self-detection of beta particles from 176Lu in one crystal and the detection of one or more prompt gamma rays detected in coincidence by the opposing crystal. The most probable coincidence detection involves the gamma rays of 202 and 307 keV, which result in two narrow photopeaks, superimposed on a continuous energy distribution due to the beta particle energy deposition. The relative intensities of the gamma ray peaks depend on crystal size and detector separation distance, as is explained by the analytical model and verified through the Monte Carlo simulations and experiments. CONCLUSIONS: The analytical model used in this work accurately explains the general features of the coincidence energy spectrum due to the presence of 176Lu in the scintillation crystals, as observed experimentally and with Monte Carlo simulations. This work will be useful to those research studies aimed at using the intrinsic radioactivity of LYSO crystals for transmission scans and detector calibration in coincidence mode.

Efficient system modeling for a small animal PET scanner with tapered DOI detectors.

A prototype small animal positron emission tomography (PET) scanner for mouse brain imaging has been developed at UC Davis. The new scanner uses tapered detector arrays with depth of interaction (DOI) measurement. In this paper, we present an efficient system model for the tapered PET scanner using matrix factorization and a virtual scanner geometry. The factored system matrix mainly consists of two components: a sinogram blurring matrix and a geometrical matrix. The geometric matrix is based on a virtual scanner geometry. The sinogram blurring matrix is estimated by matrix factorization. We investigate the performance of different virtual scanner geometries. Both simulation study and real data experiments are performed in the fully 3D mode to study the image quality under different system models. The results indicate that the proposed matrix factorization can maintain image quality while substantially reduce the image reconstruction time and system matrix storage cost. The proposed method can be also applied to other PET scanners with DOI measurement.

Two flat-backed polydesmidan millipedes from the Miocene Chiapas-amber Lagerstätte, Mexico.

Two species of fossil polydesmidan millipedes (Diplopoda: Polydesmida) embedded in amber are described from Miocene strata near Simojovel, in the Chiapas Highlands, Mexico. Maatidesmus paachtun gen. et sp. nov., placed into Chelodesmidae Cook, 1895, and Anbarrhacus adamantis gen. et sp. nov., assigned in the family Platyrhacidae Pocock, 1895. Morphological data from fossil specimens have been recovered using 3D X-ray micro-computed tomography and regular to infrared-reflected microscopy. Both fossil species are recognizable as new primarily but not exclusively, by collum margin modification and remarkable paranotal and metatergite dorsal sculpture.

A Monte Carlo investigation of the spatial resolution performance of a small-animal PET scanner designed for mouse brain imaging studies.

Our laboratory has developed PET detectors with depth-encoding accuracy of ∼2 mm based on finely pixelated crystals with a tapered geometry, readout at both ends with position-sensitive avalanche photodiodes (PSAPDs). These detectors are currently being used in our laboratory to build a one-ring high resolution PET scanner for mouse brain imaging studies. Due to the inactive areas around the PSAPDs, large gaps exist between the detector modules which can degrade the image spatial resolution obtained using analytical reconstruction with filtered backprojection (FBP). In this work, the Geant4-based GATE Monte Carlo package was used to assist in determining whether gantry rotation was necessary and to assess the expected spatial resolution of the system. The following factors were investigated: rotating vs. static gantry modes with and without compensation of missing data using the discrete cosine transform (DCT) method, two levels of depth-encoding, and positron annihilation effects for (18)F. Our results indicate that while the static scanner produces poor quality FBP images with streak and ring artifacts, the image quality was greatly improved after compensation of missing data. The simulation indicates that the expected FWHM system spatial resolution is 0.70 ± 0.05 mm, which approaches the predicted limit of 0.5 mm FWHM due to positron range, photon non-colinearity and physical detector element size effects. We conclude that excellent reconstructed resolution without gantry rotation is possible even using FBP if the gaps are appropriately handled and that this design can approach the resolution limits set by positron annihilation physics.

Small photon beam measurements using radiochromic film and Monte Carlo simulations in a water phantom.

This work reports the use of both GafChromic EBT film immersed in a water phantom and Monte Carlo (MC) simulations for small photon beam stereotactic radiosurgery dosimetry. Circularly collimated photon beams with diameters in the 4-20 mm range of a dedicated 6 MV linear accelerator (Novalis, BrainLAB, Germany) were used to perform off-axis ratios, tissue maximum ratios and total scatter factors measurements, and MC simulations. GafChromic EBT film data show an excellent agreement with MC results (<2.7%) for all measured quantities.

Radiation transmission, leakage and beam penumbra measurements of a micro-multileaf collimator using GafChromic EBT film.

Micro-multileaf collimator systems coupled to linear accelerators for radioneurosurgery treatments require a rigorous dosimetric characterization in order to be used in 3D conformal and intensity modulated stereotactic radiosurgery and radiotherapy applications. This characterization involves high precision measurements of leaf transmission, leakage and beam penumbra through the collimation system and requires the use of detectors with high spatial resolution, high sensitivity and practically no energy dependence. In this work the use of GafChromic EBT radiochromic film to measure the basic dosimetric properties of the m3-mMLC (BrainLAB, Germany) micro-multileaf collimator system integrated to a 6 MV linear accelerator, is reported. Results show that average values of transmission and leakage radiation are 0.93 +/- 0.05% and 1.08 +/- 0.08%, respectively. The 80-20% beam penumbra were found to be 2.26 +/- 0.11 mm along the leaf side (perpendicular to leaf motion) and 2.31 +/- 0.11 mm along the leaf end (parallel to leaf motion) using square field sizes ranging from 9.1 to 1.8 cm. These measurements are in agreement with values reported in the literature for the same type of mMLC using different radiation detectors.

Spatial dose distributions in solid tumors from 186Re transported by liposomes using HS radiochromic media.

PURPOSE: A procedure for the measurement of spatial dose rate distribution of beta particles emitted by 186Re-liposomes in tumoral tissue, using HS GafChromic films, is presented. METHODS: HNSCC xenografts were intratumorally injected with 3.7 or 11.1 MBq of 186Re-liposomes, and planar gamma camera images were acquired to determine the liposome retention in the tumor. After imaging, rats were sacrificed and tumors were excised and processed in slices; HS film sections were placed between slices and the tumor lobe was reassembled. Tumors and films were kept in the dark at 4 degrees C for 18 h. After irradiation, films were removed and response was read using a transmission scanner. Films were analyzed to determine two-dimensional spatial dose rate distributions and cumulative dose volume histograms. Dose rate distributions were quantified using a 60Co calibration curve, the 186Re physical half-life, and a perturbation factor that takes into account the effect of the film protective layer. RESULTS: Dose rate distributions are highly heterogeneous with maximal dose rates about 0.4 Gy h(-1) in tumors injected with 3.7 MBq and 1.3 Gy h(-1) in tumors injected with 11.1 MBq. Dose volume histograms showed dose distributed in more than 95% and 80% of the tumor when injected with the lower and the higher activity, respectively. CONCLUSION: The described procedures and techniques have shown the potential and utility of HS GafChromic film for determination of dose rate distributions in solid tumors injected intratumorally with 186Re-liposomes. The film's structure and the liposomes' biodistribution must be taken into account to obtain quantitative dose measurements.

Ionisation density dependence of the optically stimulated luminescence dose-response of AL2O3:C to low-energy charged particles.

The optically stimulated luminescence (OSL) response of Al2O3:C to high doses of gamma or beta irradiation can be used to predict the response of this material to charged particles as a function of particle fluence, particle energy and/or linear energy transfer (LET). In particular, it is predicted that track interaction effects at high particle fluences should result in linear-sublinear growth of the OSL signal. Similar considerations also predict a dependence of the fluence at which sublinearity starts upon the energy of the particles. In this work the OSL response of Al2O3:C to low-energy charged particles was investigated using protons (1, 2 and 4 MeV), carbon ions (13 MeV) and oxygen ions (10 MeV). The sublinear growth predicted above was qualitatively confirmed, but the energy dependence prediction was not. Furthermore, the efficiency of OSL production in the material after charged particle irradiation, compared to that obtained for gamma irradiation, is determined from the dose-response curves by fitting to a simple saturating exponential function. The efficiency values so obtained using this method are compared with those obtained from a conventional single-point measurement in the linear portion of the curve and found to be in good agreement. In general, the efficiency decreases as the LET of the particle increases. The present data are compared with published data obtained using high-energy charged particles and the results show that the efficiency is not a unique function of LET.