High Pitch Helical CT Reconstruction

Reducing radiation dose for NN-based COVID-19 detection in helical chest CT using real-time monitored reconstruction

Computed tomography is a powerful tool for medical examination, which plays a particularly important role in the investigation of acute diseases, such as COVID-19. A growing concern in relation to CT scans is the radiation to which the patients are exposed, and a lot of research is dedicated to methods and approaches to how to reduce the radiation dose in X-ray CT studies. In this paper, we propose a novel scanning protocol based on real-time monitored reconstruction for a helical chest CT using a pre-trained neural network model for COVID-19 detection as an expert. In a simulated study, for the first time, we proposed using per-slice stopping rules based on the COVID-19 detection neural network output to reduce the frequency of projection acquisition for portions of the scanning process. The proposed method allows reducing the total number of X-ray projections necessary for COVID-19 detection, and thus reducing the radiation dose, without a significant decrease in the prediction accuracy. The proposed protocol was evaluated on 163 patients from the COVID-CTset dataset, providing a mean dose reduction of 15.1% while the mean decrease in prediction accuracy amounted to only 1.9% achieving a Pareto improvement over a fixed protocol.

Low-density tissue scaffold imaging by synchrotron radiation propagation-based imaging computed tomography with helical acquisition mode

Visualization of low-density tissue scaffolds made from hydrogels is important yet challenging in tissue engineering and regenerative medicine (TERM). For this, synchrotron radiation propagation-based imaging computed tomography (SR-PBI-CT) has great potential, but is limited due to the ring artifacts commonly observed in SR-PBI-CT images. To address this issue, this study focuses on the integration of SR-PBI-CT and helical acquisition mode (i.e. SR-PBI-HCT) to visualize hydrogel scaffolds. The influence of key imaging parameters on the image quality of hydrogel scaffolds was investigated, including the helical pitch (p), photon energy (E) and the number of acquisition projections per rotation/revolution (N_p), and, on this basis, those parameters were optimized to improve image quality and to reduce noise level and artifacts. The results illustrate that SR-PBI-HCT imaging shows impressive advantages in avoiding ring artifacts with p = 1.5, E = 30 keV and N_p = 500 for the visualization of hydrogel scaffolds in vitro. Furthermore, the results also demonstrate that hydrogel scaffolds can be visualized using SR-PBI-HCT with good contrast while at a low radiation dose, i.e. 342 mGy (voxel size of 26 µm, suitable for in vivo imaging). This paper presents a systematic study on hydrogel scaffold imaging using SR-PBI-HCT and the results reveal that SR-PBI-HCT is a powerful tool for visualizing and characterizing low-density scaffolds with a high image quality in vitro. This work represents a significant advance toward the non-invasive in vivo visualization and characterization of hydrogel scaffolds at a suitable radiation dose.

Reconstruction of three-dimensional tomographic patient models for radiation dose modulation in CT from two scout views using deep learning

BACKGROUND

A tomographic patient model is essential for radiation dose modulation in x-ray computed tomography (CT). Currently, two-view scout images (also known as topograms) are used to estimate patient models with relatively uniform attenuation coefficients. These patient models do not account for the detailed anatomical variations of human subjects, and thus, may limit the accuracy of intraview or organ-specific dose modulations in emerging CT technologies.

PURPOSE

The purpose of this work was to show that 3D tomographic patient models can be generated from two-view scout images using deep learning strategies, and the reconstructed 3D patient models indeed enable accurate prescriptions of fluence-field modulated or organ-specific dose delivery in the subsequent CT scans.

METHODS

CT images and the corresponding two-view scout images were retrospectively collected from 4214 individual CT exams. The collected data were curated for the training of a deep neural network architecture termed ScoutCT-NET to generate 3D tomographic attenuation models from two-view scout images. The trained network was validated using a cohort of 55 136 images from 212 individual patients. To evaluate the accuracy of the reconstructed 3D patient models, radiation delivery plans were generated using ScoutCT-NET 3D patient models and compared with plans prescribed based on true CT images (gold standard) for both fluence-field-modulated CT and organ-specific CT. Radiation dose distributions were estimated using Monte Carlo simulations and were quantitatively evaluated using the Gamma analysis method. Modulated dose profiles were compared against state-of-the-art tube current modulation schemes. Impacts of ScoutCT-NET patient model-based dose modulation schemes on universal-purpose CT acquisitions and organ-specific acquisitions were also compared in terms of overall image appearance, noise magnitude, and noise uniformity.

RESULTS

The results demonstrate that (1) The end-to-end trained ScoutCT-NET can be used to generate 3D patient attenuation models and demonstrate empirical generalizability. (2) The 3D patient models can be used to accurately estimate the spatial distribution of radiation dose delivered by standard helical CTs prior to the actual CT acquisition; compared to the gold-standard dose distribution, 95.0% of the voxels in the ScoutCT-NET based dose maps have acceptable gamma values for 5 mm distance-to-agreement and 10% dose difference. (3) The 3D patient models also enabled accurate prescription of fluence-field modulated CT to generate a more uniform noise distribution across the patient body compared to tube current-modulated CT. (4) ScoutCT-NET 3D patient models enabled accurate prescription of organ-specific CT to boost image quality for a given body region-of-interest under a given radiation dose constraint.

CONCLUSION

3D tomographic attenuation models generated by ScoutCT-NET from two-view scout images can be used to prescribe fluence-field-modulated or organ-specific CT scans with high accuracy for the overall objective of radiation dose reduction or image quality improvement for a given imaging task.

Convex optimization algorithms in medical image reconstruction - in the age of AI

The past decade has seen the flourish of model based image reconstruction (MBIR) algorithms, which are often applications or adaptations of convex optimization algorithms from the optimization community. We review some state-of-the-art algorithms that have enjoyed wide popularity in medical image reconstruction, emphasize known connections between different algorithms, and discuss practical issues such as computation and memory cost. More recently, deep learning (DL) has forayed into medical imaging, where the latest development tries to exploit the synergy between DL and MBIR to elevate the MBIR's performance. We present existing approaches and emerging trends in DL-enhanced MBIR methods, with particular attention to the underlying role of convexity and convex algorithms on network architecture. We also discuss how convexity can be employed to improve the generalizability and representation power of DL networks in general.

Patient-specific hyperparameter learning for optimization-based CT image reconstruction

We propose a hyperparameter learning framework that learnspatient-specifichyperparameters for optimization-based image reconstruction problems for x-ray CT applications. The framework consists of two functional modules: (1) a hyperparameter learning module parameterized by a convolutional neural network, (2) an image reconstruction module that takes as inputs both the noisy sinogram and the hyperparameters from (1) and generates the reconstructed images. As a proof-of-concept study, in this work we focus on a subclass of optimization-based image reconstruction problems with exactly computable solutions so that the whole network can be trained end-to-end in an efficient manner. Unlike existing hyperparameter learning methods, our proposed framework generates patient-specific hyperparameters from the sinogram of the same patient. Numerical studies demonstrate the effectiveness of our proposed approach compared to bi-level optimization.