We present the ASTRA Toolbox as an open platform for 3D image reconstruction in tomography. Most of the software tools that are currently used in electron tomography offer limited flexibility with respect to the geometrical parameters of the acquisition model and the algorithms used for reconstruction. The ASTRA Toolbox provides an extensive set of fast and flexible building blocks that can be used to develop advanced reconstruction algorithms, effectively removing these limitations. We demonstrate this flexibility, the resulting reconstruction quality, and the computational efficiency of this toolbox by a series of experiments, based on experimental dual-axis tilt series.
Object reconstruction from a series of projection images, such as in computed tomography (CT), is a popular tool in many different application fields. Existing commercial software typically provides sufficiently accurate and convenient-to-use reconstruction tools to the end-user. However, in applications where a non-standard acquisition protocol is used, or where advanced reconstruction methods are required, the standard software tools often are incapable of computing accurate reconstruction images. This article introduces the ASTRA Toolbox. Aimed at researchers across multiple tomographic application fields, the ASTRA Toolbox provides a highly efficient and highly flexible open source set of tools for tomographic projection and reconstruction. The main features of the ASTRA Toolbox are discussed and several use cases are presented.
The three-dimensional (3D) atomic structure of nanomaterials, including strain, is crucial to understand their properties. Here, we investigate lattice strain in Au nanodecahedra using electron tomography. Although different electron tomography techniques enabled 3D characterizations of nanostructures at the atomic level, a reliable determination of lattice strain is not straightforward. We therefore propose a novel model-based approach from which atomic coordinates are measured. Our findings demonstrate the importance of investigating lattice strain in 3D.
In X-ray imaging, it is common practice to normalize the acquired projection data with averaged flat fields taken prior to the scan. Unfortunately, due to source instabilities, vibrating beamline components such as the monochromator, time varying detector properties, or other confounding factors, flat fields are often far from stationary, resulting in significant systematic errors in intensity normalization. In this work, a simple and efficient method is proposed to account for dynamically varying flat fields. Through principal component analysis of a set of flat fields, eigen flat fields are computed. A linear combination of the most important eigen flat fields is then used to individually normalize each Xray projection. Experiments show that the proposed dynamic flat field correction leads to a substantial reduction of systematic errors in projection intensity normalization compared to conventional flat field correction. DOI Abstract:In X-ray imaging, it is common practice to normalize the acquired projection data with averaged flat fields taken prior to the scan. Unfortunately, due to source instabilities, vibrating beamline components such as the monochromator, time varying detector properties, or other confounding factors, flat fields are often far from stationary, resulting in significant systematic errors in intensity normalization. In this work, a simple and efficient method is proposed to account for dynamically varying flat fields. Through principal component analysis of a set of flat fields, eigen flat fields are computed. A linear combination of the most important eigen flat fields is then used to individually normalize each X-ray projection. Experiments show that the proposed dynamic flat field correction leads to a substantial reduction of systematic errors in projection intensity normalization compared to conventional flat field correction.
The results indicated that the proposed framework could be used for this PET scanner with improved image quality. This method could also be applied to other state-of-the-art whole-body PET scanners and preclinical PET scanners with a similar shape.
I-131 is a frequently used isotope for radionuclide therapy. This technique for cancer treatment requires a pre-therapeutic dosimetric study. The latter is usually performed (for this radionuclide) by directly imaging the uptake of the therapeutic radionuclide in the body or by replacing it by one of its isotopes, which are more suitable for imaging. This study aimed to compare the image quality that can be achieved by three iodine isotopes: I-131 and I-123 for single-photon emission computed tomography imaging, and I-124 for positron emission tomography imaging. The imaging characteristics of each isotope were investigated by simulated data. Their spectrums, point-spread functions, and contrast-recovery curves were drawn and compared. I-131 was imaged with a high-energy all-purpose (HEAP) collimator, whereas two collimators were compared for I-123: low-energy high-resolution (LEHR) and medium energy (ME). No mechanical collimation was used for I-124. The influence of small high-energy peaks (>0.1%) on the main energy window contamination were evaluated. Furthermore, the effect of a scattering medium was investigated and the triple energy window (TEW) correction was used for spectral-based scatter correction. Results showed that I-123 gave the best results with a LEHR collimator when the scatter correction was applied. Without correction, the ME collimator reduced the effects of high-energy contamination. I-131 offered the worst results. This can be explained by the large amount of septal penetration from the photopeak and by the collimator, which gave a low spatial resolution. I-124 gave the best imaging properties owing to its electronic collimation (high sensitivity) and a short coincidence time window.
4D computed tomography (4D-CT) aims to visualise the temporal dynamics of a 3D sample with a sufficiently high temporal and spatial resolution. Successive time frames are typically obtained by sequential scanning, followed by independent reconstruction of each 3D dataset. Such an approach requires a large number of projections for each scan to obtain images with sufficient quality (in terms of artefacts and SNR). Hence, there is a clear trade-off between the rotation speed of the gantry (i.e. time resolution) and the quality of the reconstructed images. In this paper, the MotionVector-based Iterative Technique (MoVIT) is introduced which reconstructs a particular time frame by including the projections of neighbouring time frames as well. It is shown that such a strategy improves the trade-off between the rotation speed and the SNR. The framework is tested on both numerical simulations and on 4D X-ray CT datasets of polyurethane foam under compression. Results show that reconstructions obtained with MoVIT have a significantly higher SNR compared to the SNR of conventional 4D reconstructions.
Today, new single photon emission computed tomography (SPECT) reconstruction techniques rely on accurate Monte Carlo (MC) simulations to optimize reconstructed images. However, existing MC scintillation camera models which usually include an accurate description of the collimator and crystal, lack correct implementation of the gamma camera's back compartments. In the case of dual isotope simultaneous acquisition (DISA), where backscattered photons from the highest energy isotope are detected in the imaging energy window of the second isotope, this approximation may induce simulation errors. Here, we investigate the influence of backscatter compartment modelling on the simulation accuracy of high-energy isotopes. Three models of a scintillation camera were simulated: a simple model (SM), composed only of a collimator and a NaI(Tl) crystal; an intermediate model (IM), adding a simplified description of the backscatter compartments to the previous model and a complete model (CM), accurately simulating the materials and geometries of the camera. The camera models were evaluated with point sources ((67)Ga, (99m)Tc, (111)In, (123)I, (131)I and (18)F) in air without a collimator, in air with a collimator and in water with a collimator. In the latter case, sensitivities and point-spread functions (PSFs) simulated in the photopeak window with the IM and CM are close to the measured values (error below 10.5%). In the backscatter energy window, however, the IM and CM overestimate the FWHM of the detected PSF by 52% and 23%, respectively, while the SM underestimates it by 34%. The backscatter peak fluence is also overestimated by 20% and 10% with the IM and CM, respectively, whereas it is underestimated by 60% with the SM. The results show that an accurate description of the backscatter compartments is required for SPECT simulations of high-energy isotopes (above 300 keV) when the backscatter energy window is of interest.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.