XCIST-an open access x-ray/CT simulation toolkit

Artificial Intelligence in Computed Tomography Image Reconstruction: A Review of Recent Advances

The development of novel image reconstruction algorithms has been pivotal in enhancing image quality and reducing radiation dose in computed tomography (CT) imaging. Traditional techniques like filtered back projection perform well under ideal conditions but fail to generate high-quality images under low-dose, sparse-view, and limited-angle conditions. Iterative reconstruction methods improve upon filtered back projection by incorporating system models and assumptions about the patient, yet they can suffer from patchy image textures. The emergence of artificial intelligence (AI), particularly deep learning, has further advanced CT reconstruction. AI techniques have demonstrated great potential in reducing radiation dose while preserving image quality and noise texture. Moreover, AI has exhibited unprecedented performance in addressing challenging CT reconstruction problems, including low-dose CT, sparse-view CT, limited-angle CT, and interior tomography. This review focuses on the latest advances in AI-based CT reconstruction under these challenging conditions.

Toward widespread use of virtual trials in medical imaging innovation and regulatory science

The rapid advancement in the field of medical imaging presents a challenge in keeping up to date with the necessary objective evaluations and optimizations for safe and effective use in clinical settings. These evaluations are traditionally done using clinical imaging trials, which while effective, pose several limitations including high costs, ethical considerations for repetitive experiments, time constraints, and lack of ground truth. To tackle these issues, virtual trials (aka in silico trials) have emerged as a promising alternative, using computational models of human subjects and imaging devices, and observer models/analysis to carry out experiments. To facilitate the widespread use of virtual trials within the medical imaging research community, a major need is to establish a common consensus framework that all can use. Based on the ongoing efforts of an AAPM Task Group (TG387), this article provides a comprehensive overview of the requirements for establishing virtual imaging trial frameworks, paving the way toward their widespread use within the medical imaging research community. These requirements include credibility, reproducibility, and accessibility. Credibility assessment involves verification, validation, uncertainty quantification, and sensitivity analysis, ensuring the accuracy and realism of computational models. A proper credibility assessment requires a clear context of use and the questions that the study is intended to objectively answer. For reproducibility and accessibility, this article highlights the need for detailed documentation, user-friendly software packages, and standard input/output formats. Challenges in data and software sharing, including proprietary data and inconsistent file formats, are discussed. Recommended solutions to enhance accessibility include containerized environments and data-sharing hubs, along with following standards such as CDISC (Clinical Data Interchange Standards Consortium). By addressing challenges associated with credibility, reproducibility, and accessibility, virtual imaging trials can be positioned as a powerful and inclusive resource, advancing medical imaging innovation and regulatory science.

Metric learning guided sinogram denoising for cone beam CT enhancement

BACKGROUND

Cone beam computed tomography (CBCT) is a widely available modality, but its clinical utility has been limited by low detail conspicuity and quantitative accuracy. Convenient post-reconstruction denoising is subject to back projected patterned residual, but joint denoise-reconstruction is typically computationally expensive and complex.

PURPOSE

In this study, we develop and evaluate a novel Metric-learning guided wavelet transform reconstruction (MEGATRON) approach to enhance image domain quality with projection-domain processing.

METHODS

Projection domain based processing has the benefit of being simple, efficient, and compatible with various reconstruction toolkit and vendor platforms. However, they also typically show inferior performance in the final reconstructed image, because the denoising goals in projection and image domains do not necessarily align. Motivated by these observations, this work aims to translate the demand for quality enhancement from the quantitative image domain to the more easily operable projection domain. Specifically, the proposed paradigm consists of a metric learning module and a denoising network module. Via metric learning, enhancement objectives on the wavelet encoded sinogram domain data are defined to reflect post-reconstruction image discrepancy. The denoising network maps measured cone-beam projection to its enhanced version, driven by the learnt objective. In doing so, the denoiser operates in the convenient sinogram to sinogram fashion but reflects improvement in reconstructed image as the final goal. Implementation-wise, metric learning was formalized as optimizing the weighted fitting of wavelet subbands, and a res-Unet, which is a Unet structure with residual blocks, was used for denoising. To access quantitative reference, cone-beam projections were simulated using the X-ray based Cancer Imaging Simulation Toolkit (XCIST). In both learning modules, a data set of 123 human thoraxes, which was from Open-Source Imaging Consortium (OSIC) Pulmonary Fibrosis Progression challenge, was used. Reconstructed CBCT thoracic images were compared against ground truth FB and performance was assessed in root mean square error (RMSE), peak signal-to-noise ratio (PSNR), and structural similarity index (SSIM).

RESULTS

MEGATRON achieved RMSE in HU value, PSNR, and SSIM were 30.97 ± 4.25, 37.45 ± 1.78, and 93.23 ± 1.62, respectively. These values are on par with reported results from sophisticated physics-driven CBCT enhancement, demonstrating promise and utility of the proposed MEGATRON method.

CONCLUSION

We have demonstrated that incorporating the proposed metric learning into sinogram denoising introduces awareness of reconstruction goal and improves final quantitative performance. The proposed approach is compatible with a wide range of denoiser network structures and reconstruction modules, to suit customized need or further improve performance.

Coronary atherosclerotic plaque characterization with silicon-based photon-counting computed tomography (CT): A simulation-based feasibility study

BACKGROUND

Recent photon-counting computed tomography (PCCT) development brings great opportunities for plaque characterization with much-improved spatial resolution and spectral imaging capability. While existing coronary plaque PCCT imaging results are based on CZT- or CdTe-materials detectors, deep-silicon photon-counting detectors offer unique performance characteristics and promise distinct imaging capabilities.

PURPOSE

This study aims to numerically investigate the feasibility of characterizing plaques with a deep-silicon PCCT scanner and to demonstrate its potential performance advantages over traditional CT scanners using energy-integrating detectors (EID).

METHODS

We conducted a systematic simulation study of a deep-silicon PCCT scanner using a newly developed digital plaque phantom with clinically relevant geometrical and chemical properties. Through qualitative and quantitative evaluations, this study investigates the effects of spatial resolution, noise, and motion artifacts on plaque imaging.

RESULTS

Noise-free simulations indicated that PCCT imaging could delineate the boundary of necrotic cores with a much finer resolution than EID-CT imaging, achieving a structural similarity index metric (SSIM) score of 0.970 and reducing the root mean squared error (RMSE) by two-thirds. Measuring necrotic core area errors were reduced from 91.5% to 24%, and fibrous cap thickness errors were reduced from 349.8% to 33.3%. In the presence of noise, the optimal reconstruction was achieved using 0.25 mm voxels and a soft reconstruction kernel, yielding the highest contrast-to-noise ratio (CNR) of 3.48 for necrotic core detection and the best image quality metrics among all choices. However, the ultrahigh resolution of PCCT increased sensitivity to motion artifacts, which could be mitigated by keeping residual motion amplitude below 0.4 mm.

CONCLUSIONS

The findings suggest that deep-silicon PCCT scanner can offer sufficient spatial resolution and tissue contrast for effective plaque characterization, potentially improving diagnostic accuracy in cardiovascular imaging, provided image noise and motion blur can be mitigated using advanced algorithms. This simulation study involves several simplifications, which may result in some idealized outcomes that do not directly translate to clinical practice. Further validation studies with physical scans are necessary and will be considered for future work.

Deep learning estimation of proton stopping power with photon-counting computed tomography: a virtual study

Purpose

Proton radiation therapy may achieve precise dose delivery to the tumor while sparing non-cancerous surrounding tissue, owing to the distinct Bragg peaks of protons. Aligning the high-dose region with the tumor requires accurate estimates of the proton stopping power ratio (SPR) of patient tissues, commonly derived from computed tomography (CT) image data. Photon-counting detectors for CT have demonstrated advantages over their energy-integrating counterparts, such as improved quantitative imaging, higher spatial resolution, and filtering of electronic noise. We assessed the potential of photon-counting computed tomography (PCCT) for improving SPR estimation by training a deep neural network on a domain transform from PCCT images to SPR maps.

Approach

The XCAT phantom was used to simulate PCCT images of the head with CatSim, as well as to compute corresponding ground truth SPR maps. The tube current was set to 260 mA, tube voltage to 120 kV, and number of view angles to 4000. The CT images and SPR maps were used as input and labels for training a U-Net.

Results

Prediction of SPR with the network yielded average root mean square errors (RMSE) of 0.26% to 0.41%, which was an improvement on the RMSE for methods based on physical modeling developed for single-energy CT at 0.40% to 1.30% and dual-energy CT at 0.41% to 3.00%, performed on the simulated PCCT data.

Conclusions

These early results show promise for using a combination of PCCT and deep learning for estimating SPR, which in extension demonstrates potential for reducing the beam range uncertainty in proton therapy.

A Denoising Diffusion Probabilistic Model for Metal Artifact Reduction in CT

The presence of metal objects leads to corrupted CT projection measurements, resulting in metal artifacts in the reconstructed CT images. AI promises to offer improved solutions to estimate missing sinogram data for metal artifact reduction (MAR), as previously shown with convolutional neural networks (CNNs) and generative adversarial networks (GANs). Recently, denoising diffusion probabilistic models (DDPM) have shown great promise in image generation tasks, potentially outperforming GANs. In this study, a DDPM-based approach is proposed for inpainting of missing sinogram data for improved MAR. The proposed model is unconditionally trained, free from information on metal objects, which can potentially enhance its generalization capabilities across different types of metal implants compared to conditionally trained approaches. The performance of the proposed technique was evaluated and compared to the state-of-the-art normalized MAR (NMAR) approach as well as to CNN-based and GAN-based MAR approaches. The DDPM-based approach provided significantly higher SSIM and PSNR, as compared to NMAR (SSIM: p [Formula: see text]; PSNR: p [Formula: see text]), the CNN (SSIM: p [Formula: see text]; PSNR: p [Formula: see text]) and the GAN (SSIM: p [Formula: see text]; PSNR: p <0.05) methods. The DDPM-MAR technique was further evaluated based on clinically relevant image quality metrics on clinical CT images with virtually introduced metal objects and metal artifacts, demonstrating superior quality relative to the other three models. In general, the AI-based techniques showed improved MAR performance compared to the non-AI-based NMAR approach. The proposed methodology shows promise in enhancing the effectiveness of MAR, and therefore improving the diagnostic accuracy of CT.

b-MAR: bidirectional artifact representations learning framework for metal artifact reduction in dental CBCT

Objective.Metal artifacts in computed tomography (CT) images hinder diagnosis and treatment significantly. Specifically, dental cone-beam computed tomography (Dental CBCT) images are seriously contaminated by metal artifacts due to the widespread use of low tube voltages and the presence of various high-attenuation materials in dental structures. Existing supervised metal artifact reduction (MAR) methods mainly learn the mapping of artifact-affected images to clean images, while ignoring the modeling of the metal artifact generation process. Therefore, we propose the bidirectional artifact representations learning framework to adaptively encode metal artifacts caused by various dental implants and model the generation and elimination of metal artifacts, thereby improving MAR performance.Approach.Specifically, we introduce an efficient artifact encoder to extract multi-scale representations of metal artifacts from artifact-affected images. These extracted metal artifact representations are then bidirectionally embedded into both the metal artifact generator and the metal artifact eliminator, which can simultaneously improve the performance of artifact removal and artifact generation. The artifact eliminator learns artifact removal in a supervised manner, while the artifact generator learns artifact generation in an adversarial manner. To further improve the performance of the bidirectional task networks, we propose artifact consistency loss to align the consistency of images generated by the eliminator and the generator with or without embedding artifact representations.Main results.To validate the effectiveness of our algorithm, experiments are conducted on simulated and clinical datasets containing various dental metal morphologies. Quantitative metrics are calculated to evaluate the results of the simulation tests, which demonstrate b-MAR improvements of >1.4131 dB in PSNR, >0.3473 HU decrements in RMSE, and >0.0025 promotion in structural similarity index measurement over the current state-of-the-art MAR methods. All results indicate that the proposed b-MAR method can remove artifacts caused by various metal morphologies and restore the structural integrity of dental tissues effectively.Significance.The proposed b-MAR method strengthens the joint learning of the artifact removal process and the artifact generation process by bidirectionally embedding artifact representations, thereby improving the model's artifact removal performance. Compared with other comparison methods, b-MAR can robustly and effectively correct metal artifacts in dental CBCT images caused by different dental metals.

Intra-Guided Dual-Energy Material Decomposition Method for Dual-Source CBCT Systems

Dual-energy CBCT imaging plays a crucial role in advanced imaging applications due to its ability to quantify material components. Although there are multiple established systems for dual-energy imaging, they often come with high deployment costs. This study investigates the feasibility of dual-energy material decomposition on a less expensive dual-source CBCT system. Due to the specific geometry of dual-source CBCT, cone-beam (CB) artifacts seriously degrade the material decomposition quality when the direct inversion method is adopted. In this paper, a novel dual-source intra-guided material decomposition framework is proposed to suppress CB artifacts in the material maps. Firstly, a dual-source two-pass algorithm is developed for artifact pre-correction during image reconstruction. Then, the structure of small cone-angle regions from one X-ray source is utilized to guide cone beam artifact removal in large cone-angle regions from another source. Finally, direct matrix inverse material decomposition is applied to the filtered image. Extensive experiments have been conducted on the authentic human phantom and demonstrate that the proposed framework achieves excellent performance in cone beam artifact reduction compared to other methods, and significantly enhances the quality of the material maps.

Simulated deep CT characterization of liver metastases with high-resolution filtered back projection reconstruction

Early diagnosis and accurate prognosis of colorectal cancer is critical for determining optimal treatment plans and maximizing patient outcomes, especially as the disease progresses into liver metastases. Computed tomography (CT) is a frontline tool for this task; however, the preservation of predictive radiomic features is highly dependent on the scanning protocol and reconstruction algorithm. We hypothesized that image reconstruction with a high-frequency kernel could result in a better characterization of liver metastases features via deep neural networks. This kernel produces images that appear noisier but preserve more sinogram information. A simulation pipeline was developed to study the effects of imaging parameters on the ability to characterize the features of liver metastases. This pipeline utilizes a fractal approach to generate a diverse population of shapes representing virtual metastases, and then it superimposes them on a realistic CT liver region to perform a virtual CT scan using CatSim. Datasets of 10,000 liver metastases were generated, scanned, and reconstructed using either standard or high-frequency kernels. These data were used to train and validate deep neural networks to recover crafted metastases characteristics, such as internal heterogeneity, edge sharpness, and edge fractal dimension. In the absence of noise, models scored, on average, 12.2% ( α = 0.012 ) and 7.5% ( α = 0.049 ) lower squared error for characterizing edge sharpness and fractal dimension, respectively, when using high-frequency reconstructions compared to standard. However, the differences in performance were statistically insignificant when a typical level of CT noise was simulated in the clinical scan. Our results suggest that high-frequency reconstruction kernels can better preserve information for downstream artificial intelligence-based radiomic characterization, provided that noise is limited. Future work should investigate the information-preserving kernels in datasets with clinical labels.

Fine structural human phantom in dentistry and instance tooth segmentation

In this study, we present the development of a fine structural human phantom designed specifically for applications in dentistry. This research focused on assessing the viability of applying medical computer vision techniques to the task of segmenting individual teeth within a phantom. Using a virtual cone-beam computed tomography (CBCT) system, we generated over 170,000 training datasets. These datasets were produced by varying the elemental densities and tooth sizes within the human phantom, as well as varying the X-ray spectrum, noise intensity, and projection cutoff intensity in the virtual CBCT system. The deep-learning (DL) based tooth segmentation model was trained using the generated datasets. The results demonstrate an agreement with manual contouring when applied to clinical CBCT data. Specifically, the Dice similarity coefficient exceeded 0.87, indicating the robust performance of the developed segmentation model even when virtual imaging was used. The present results show the practical utility of virtual imaging techniques in dentistry and highlight the potential of medical computer vision for enhancing precision and efficiency in dental imaging processes.

Development and tuning of models for accurate simulation of CT spatial resolution using CatSim

Objective. We sought to systematically evaluate CatSim's ability to accurately simulate the spatial resolution produced by a typical 64-detector-row clinical CT scanner in the projection and image domains, over the range of clinically used x-ray techniques.Approach.Using a 64-detector-row clinical scanner, we scanned two phantoms designed to evaluate spatial resolution in the projection and image domains. These empirical scans were performed over the standard clinically used range of x-ray techniques (kV, and mA). We extracted projection data from the scanner, and we reconstructed images. For the CatSim simulations, we developed digital phantoms to represent the phantoms used in the empirical scans. We developed a new, realistic model for the x-ray source focal spot, and we empirically tuned a published model for the x-ray detector temporal response. We applied these phantoms and models to simulate scans equivalent to the empirical scans, and we reconstructed the simulated projections using the same methods used for the empirical scans. For the empirical and simulated scans, we qualitatively and quantitatively compared the projection-domain and image-domain point-spread functions (PSFs) as well as the image-domain modulation transfer functions. We reported four quantitative metrics and the percent error between the empirical and simulated results.Main Results.Qualitatively, the PSFs matched well in both the projection and image domains. Quantitatively, all four metrics generally agreed well, with most of the average errors substantially less than 5% for all x-ray techniques. Although the errors tended to increase with decreasing kV, we found that the CatSim simulations agreed with the empirical scans within limits required for the anticipated applications of CatSim.Significance.The new focal spot model and the new detector temporal response model are significant contributions to CatSim because they enabled achieving the desired level of agreement between empirical and simulated results. With these new models and this validation, CatSim users can be confident that the spatial resolution represented by simulations faithfully represents results that would be obtained by a real scanner, within reasonable, known limits. Furthermore, users of CatSim can vary parameters including but not limited to system geometry, focal spot size/shape and detector parameters, beyond the values available in physical scanners, and be confident in the results. Therefore, CatSim can be used to explore new hardware designs as well as new scanning and reconstruction methods, thus enabling acceleration of improved CT scan capabilities.

Deep-silicon photon-counting x-ray projection denoising through reinforcement learning

BACKGROUND

In recent years, deep reinforcement learning (RL) has been applied to various medical tasks and produced encouraging results.

OBJECTIVE

In this paper, we demonstrate the feasibility of deep RL for denoising simulated deep-silicon photon-counting CT (PCCT) data in both full and interior scan modes. PCCT offers higher spatial and spectral resolution than conventional CT, requiring advanced denoising methods to suppress noise increase.

METHODS

In this work, we apply a dueling double deep Q network (DDDQN) to denoise PCCT data for maximum contrast-to-noise ratio (CNR) and a multi-agent approach to handle data non-stationarity.

RESULTS

Using our method, we obtained significant image quality improvement for single-channel scans and consistent improvement for all three channels of multichannel scans. For the single-channel interior scans, the PSNR (dB) and SSIM increased from 33.4078 and 0.9165 to 37.4167 and 0.9790 respectively. For the multichannel interior scans, the channel-wise PSNR (dB) increased from 31.2348, 30.7114, and 30.4667 to 31.6182, 30.9783, and 30.8427 respectively. Similarly, the SSIM improved from 0.9415, 0.9445, and 0.9336 to 0.9504, 0.9493, and 0.0326 respectively.

CONCLUSIONS

Our results show that the RL approach improves image quality effectively, efficiently, and consistently across multiple spectral channels and has great potential in clinical applications.

PDS-MAR: a fine-grained projection-domain segmentation-based metal artifact reduction method for intraoperative CBCT images with guidewires

Objective.Since the invention of modern Computed Tomography (CT) systems, metal artifacts have been a persistent problem. Due to increased scattering, amplified noise, and limited-angle projection data collection, it is more difficult to suppress metal artifacts in cone-beam CT, limiting its use in human- and robot-assisted spine surgeries where metallic guidewires and screws are commonly used.Approach.To solve this problem, we present a fine-grained projection-domain segmentation-based metal artifact reduction (MAR) method termed PDS-MAR, in which metal traces are augmented and segmented in the projection domain before being inpainted using triangular interpolation. In addition, a metal reconstruction phase is proposed to restore metal areas in the image domain.Main results.The proposed method is tested on both digital phantom data and real scanned cone-beam computed tomography (CBCT) data. It achieves much-improved quantitative results in both metal segmentation and artifact reduction in our phantom study. The results on real scanned data also show the superiority of this method.Significance.The concept of projection-domain metal segmentation would advance MAR techniques in CBCT and has the potential to push forward the use of intraoperative CBCT in human-handed and robotic-assisted minimal invasive spine surgeries.

Virtual cone-beam computed tomography simulator with human phantom library and its application to the elemental material decomposition

PURPOSE

The purpose of this study is to develop a virtual CBCT simulator with a head and neck (HN) human phantom library and to demonstrate the feasibility of elemental material decomposition (EMD) for quantitative CBCT imaging using this virtual simulator.

METHODS

The library of 36 HN human phantoms were developed by extending the ICRP 110 adult phantoms based on human age, height, and weight statistics. To create the CBCT database for the library, a virtual CBCT simulator that simulated the direct and scattered X-ray on a flat panel detector using ray-tracing and deep-learning (DL) models was used. Gaussian distributed noise was also included on the flat panel detector, which was evaluated using a real CBCT system. The usefulness of the virtual CBCT system was demonstrated through the application of the developed DL-based EMD model for case involving virtual phantom and real patient.

RESULTS

The virtual simulator could generate various virtual CBCT images based on the human phantom library, and the prediction of the EMD could be successfully performed by preparing the CBCT database from the proposed virtual system, even for a real patient. The CBCT image degradation owing to the scattered X-ray and the statistical noise affected the prediction accuracy, although these effects were minimal. Furthermore, the elemental distribution using the real CBCT image was also predictable.

CONCLUSIONS

This study demonstrated the potential of using computer vision for medical data preparation and analysis, which could have important implications for improving patient outcomes, especially in adaptive radiation therapy.

A Dual-Energy Metal Artifact Redcution Method for DECT Image Reconstruction

Metal implants are one of the culprits for image quality degradation in CT imaging, introducing so-called metal artifacts. With the help of the virtual-monochromatic imaging technique, dual-energy CT has been proven to be effective in metal artifact reduction. However, the virtual monochromatic images with suppressed metal artifacts show reduced CNR compared to polychromatic images. To remove metal artifacts on polychromatic images, we propose a dual-energy NMAR (deNMAR) algorithm in this paper that adds material decomposition to the widely used NMAR framework. The dual energy sinograms are first decomposed into water and bone sinograms, and metal regions are replaced with water on the reconstructed material maps. Prior sinograms are constructed by polyenergetically forward projecting the material maps with corresponding spectra, and they are used to guide metal trace interpolation in the same way as in the NMAR algorithm. We performed experiments on authentic human body phantoms, and the results show that the proposed deNMAR algorithm achieves better performance in tissue restoration compared to other compelling methods. Tissue boundaries become clear around metal implants, and CNR rises to 2.58 from ~1.70 on 80 kV images compared to other dual-energy-based algorithms.