Deep learning based image reconstruction algorithm for limited-angle translational computed tomography

Machine learning assists in increasing the time resolution of X-ray computed tomography applied to mineral precipitation in porous media

Many subsurface engineering technologies or natural processes cause porous medium properties, such as porosity or permeability, to evolve in time. Studying and understanding such processes on the pore scale is strongly aided by visualizing the details of geometric and morphological changes in the pores. For realistic 3D porous media, X-Ray Computed Tomography (XRCT) is the method of choice for visualization. However, the necessary high spatial resolution requires either access to limited high-energy synchrotron facilities or data acquisition times which are considerably longer (e.g. hours) than the time scales of the processes causing the pore geometry change (e.g. minutes). Thus, so far, conventional benchtop XRCT technologies are often too slow to allow for studying dynamic processes. Interrupting experiments for performing XRCT scans is also in many instances no viable approach. We propose a novel workflow for investigating dynamic precipitation processes in porous media systems in 3D using a conventional XRCT technology. Our workflow is based on limiting the data acquisition time by reducing the number of projections and enhancing the lower-quality reconstructed images using machine-learning algorithms trained on images reconstructed from high-quality initial- and final-stage scans. We apply the proposed workflow to induced carbonate precipitation within a porous-media sample of sintered glass-beads. So we were able to increase the temporal resolution sufficiently to study the temporal evolution of the precipitate accumulation using an available benchtop XRCT device.

Noise-resilient deep learning for integrated circuit tomography

X-ray tomography is a non-destructive imaging technique that reveals the interior of an object from its projections at different angles. Under sparse-view and low-photon sampling, regularization priors are required to retrieve a high-fidelity reconstruction. Recently, deep learning has been used in X-ray tomography. The prior learned from training data replaces the general-purpose priors in iterative algorithms, achieving high-quality reconstructions with a neural network. Previous studies typically assume the noise statistics of test data are acquired a priori from training data, leaving the network susceptible to a change in the noise characteristics under practical imaging conditions. In this work, we propose a noise-resilient deep-reconstruction algorithm and apply it to integrated circuit tomography. By training the network with regularized reconstructions from a conventional algorithm, the learned prior shows strong noise resilience without the need for additional training with noisy examples, and allows us to obtain acceptable reconstructions with fewer photons in test data. The advantages of our framework may further enable low-photon tomographic imaging where long acquisition times limit the ability to acquire a large training set.

Four-Dimensional Cone Beam CT Imaging Using a Single Routine Scan via Deep Learning

A novel method is proposed to obtain four-dimensional (4D) cone-beam computed tomography (CBCT) images from a routine scan in patients with upper abdominal cancer. The projections are sorted according to the location of the lung diaphragm before being reconstructed to phase-sorted data. A multiscale-discriminator generative adversarial network (MSD-GAN) is proposed to alleviate the severe streaking artifacts in the original images. The MSD-GAN is trained using simulated CBCT datasets from patient planning CT images. The enhanced images are further used to estimate the deformable vector field (DVF) among breathing phases using a deformable image registration method. The estimated DVF is then applied in the motion-compensated ordered-subset simultaneous algebraic reconstruction approach to generate 4D CBCT images. The proposed MSD-GAN is compared with U-Net on the performance of image enhancement. Results show that the proposed method significantly outperforms the total variation regularization-based iterative reconstruction approach and the method using only MSD-GAN to enhance original phase-sorted images in simulation and patient studies on 4D reconstruction quality. The MSD-GAN also shows higher accuracy than the U-Net. The proposed method enables a practical way for 4D-CBCT imaging from a single routine scan in upper abdominal cancer treatment including liver and pancreatic tumors.

Dedicated cone-beam breast CT: Data acquisition strategies based on projection angle-dependent normalized glandular dose coefficients

BACKGROUND

Dedicated cone-beam breast computed tomography (CBBCT) using short-scan acquisition is being actively investigated to potentially reduce the radiation dose to the breast. This would require determining the optimal x-ray source trajectory for such short-scan acquisition.

PURPOSE

To quantify the projection angle-dependent normalized glandular dose coefficient (D g N C T $Dg{N^{CT}}$ ) in CBBCT, referred to as angularD g N C T $Dg{N^{CT}}$ , so that the x-ray ray source trajectory that minimizes the radiation dose to the breast for short-scan acquisition can be determined.

MATERIALS AND METHODS

A cohort of 75 CBBCT clinical datasets was segmented and used to generate three breast models - (I) patient-specific breast with heterogeneous fibroglandular tissue distribution and real breast shape, (II) patient-specific breast shape with homogeneous tissue distribution and matched fibroglandular weight fraction, and (III) homogeneous semi-ellipsoidal breast with patient-specific breast dimensions and matched fibroglandular weight fraction, which corresponds to the breast model used in current radiation dosimetry protocols. For each clinical dataset, the angularD g N C T $Dg{N^{CT}}$ was obtained at 10 discrete angles, spaced 36° apart, for full-scan, circular, x-ray source trajectory from Monte Carlo simulations. Model III is used for validating the Monte Carlo simulation results. Models II and III are used to determine if breast shape contributes to the observed trends in angularD g N C T $Dg{N^{CT}}$ . A geometry-based theory in conjunction with center-of-mass (C O M $COM$ ) based distribution analysis is used to explain the projection angle-dependent variation in angularD g N C T $Dg{N^{CT}}$ .

RESULTS

The theoretical model predicted that the angularD g N C T $Dg{N^{CT}}$ will follow a sinusoidal pattern and the amplitude of the sinusoid increases when the center-of-mass of fibroglandular tissue (C O M f $CO{M_f}$ ) is farther from the center-of-mass of the breast (C O M b $CO{M_b}$ ). It also predicted that the angularD g N C T $Dg{N^{CT}}$ will be minimized at x-ray source positions complementary to theC O M f $CO{M_f}$ . TheC O M f $CO{M_f}$ was superior to theC O M b $CO{M_b}$ in 80% (60/75) of the breasts. From Monte Carlo simulations and for homogeneous breasts (models II and III), the deviation in breast shape from a semi-ellipsoid had minimal effect on angularD g N C T $Dg{N^{CT}}$ and showed less than 4% variation. From Monte Carlo simulations and for model I, as predicted by our theory, the angularD g N C T $Dg{N^{CT}}$ followed a sinusoidal pattern with maxima and minima at x-ray source positions superior and inferior to the breast, respectively. For model I, the projection angle-dependent variation in angularD g N C T $Dg{N^{CT}}$ was 16.4%.

CONCLUSION

The heterogeneous tissue distribution affected the angularD g N C T $Dg{N^{CT}}$ more than the breast shape. For model I, the angularD g N C T $Dg{N^{CT}}$ was lowest when the x-ray source was inferior to the breast. Hence, for short-scan CBBCT acquisition withC O M b $CO{M_b}$ aligned with axis-of-rotation, an x-ray source trajectory inferior to the breast is preferable and such an acquisition spanning 205° can potentially reduce the mean glandular dose by up to 52%.

A Super-resolution Guided Network for Improving Automated Thyroid Nodule Segmentation

BACKGROUND AND OBJECTIVE

A thyroid nodule is an abnormal lump that grows in the thyroid gland, which is the early symptom of thyroid cancer. In order to diagnose and treat thyroid cancer at the earliest stage, it is desired to characterize the nodule accurately. Ultrasound thyroid nodules segmentation is a challenging task due to the speckle noise, intensity heterogeneity, low contrast and low resolution. In this paper, we propose a novel framework to improve the accuracy of thyroid nodules segmentation.

METHODS

Different from previous work, a super-resolution reconstruction network is firstly constructed to upscale the resolution of the input ultrasound image. After that, our proposed N-shape network is utilized to perform the segmentation task. The guidance of super-resolution reconstruction network can make the high-frequency information of the input thyroid ultrasound image richer and more comprehensive than the original image. Our N-shape network consists of several atrous spatial pyramid pooling blocks, a multi-scale input layer, a U-shape convolutional network with attention blocks and a proposed parallel atrous convolution(PAC) module. These modules are conducive to capture context information at multiple scales so that semantic features can be fully utilized for lesion segmentation. Especially, our proposed PAC module is beneficial to further improve the segmentation by extracting high-level semantic features from different receptive fields. We use the UTNI-2021 dataset for model training, validating and testing.

RESULTS

The experimental results show that our proposed method achieve a Dice value of 91.9%, a mIoU value of 87.0%, a Precision value of 88.0%, a Recall value 83.7% and a F1-score value of 84.3%, which outperforms most state-of-the-art methods.

CONCLUSIONS

Our method achieves the best performance on the UTNI-2021 dataset and provides a new way of ultrasound image segmentation. We believe that our method can provide doctors with reliable auxiliary diagnosis information in clinical practice.

Self-supervised next view prediction for limited-angle optical projection tomography

Optical projection tomography captures 2-D projections of rotating biological samples and computationally reconstructs 3-D structures from these projections, where hundreds of views with an angular range of π radian is desired for a reliable reconstruction. Limited-angle tomography tries to recover the structures of the sample using fewer angles of projections. However, the result is far from satisfactory due to the missing of wedge information. Here we introduce a novel view prediction technique, which is able to extending the angular range of captured views for the limited-angle tomography. Following a self-supervised technique that learns the relationship between the captured limited-angle views, unseen views can be computationally synthesized without any prior label data required. Combined with an optical tomography system, the proposed approach can robustly generate new projections of unknown biological samples and extends the angles of the projections from the original 60° to nearly 180°, thereby yielding high-quality 3-D reconstructions of samples even with highly incomplete measurement.

Wavelet subband-specific learning for low-dose computed tomography denoising

Deep neural networks have shown great improvements in low-dose computed tomography (CT) denoising. Early algorithms were primarily optimized to obtain an accurate image with low distortion between the denoised image and reference full-dose image at the cost of yielding an overly smoothed unrealistic CT image. Recent research has sought to preserve the fine details of denoised images with high perceptual quality, which has been accompanied by a decrease in objective quality due to a trade-off between perceptual quality and distortion. We pursue a network that can generate accurate and realistic CT images with high objective and perceptual quality within one network, achieving a better perception-distortion trade-off. To achieve this goal, we propose a stationary wavelet transform-assisted network employing the characteristics of high- and low-frequency domains of the wavelet transform and frequency subband-specific losses defined in the wavelet domain. We first introduce a stationary wavelet transform for the network training procedure. Then, we train the network using objective loss functions defined for high- and low-frequency domains to enhance the objective quality of the denoised CT image. With this network design, we train the network again after replacing the objective loss functions with perceptual loss functions in high- and low-frequency domains. As a result, we acquired denoised CT images with high perceptual quality using this strategy while minimizing the objective quality loss. We evaluated our algorithms on the phantom and clinical images, and the quantitative and qualitative results indicate that ours outperform the existing state-of-the-art algorithms in terms of objective and perceptual quality.

A review on the application of deep learning for CT reconstruction, bone segmentation and surgical planning in oral and maxillofacial surgery

Computer-assisted surgery (CAS) allows clinicians to personalize treatments and surgical interventions and has therefore become an increasingly popular treatment modality in maxillofacial surgery. The current maxillofacial CAS consists of three main steps: (1) CT image reconstruction, (2) bone segmentation, and (3) surgical planning. However, each of these three steps can introduce errors that can heavily affect the treatment outcome. As a consequence, tedious and time-consuming manual post-processing is often necessary to ensure that each step is performed adequately. One way to overcome this issue is by developing and implementing neural networks (NNs) within the maxillofacial CAS workflow. These learning algorithms can be trained to perform specific tasks without the need for explicitly defined rules. In recent years, an extremely large number of novel NN approaches have been proposed for a wide variety of applications, which makes it a difficult task to keep up with all relevant developments. This study therefore aimed to summarize and review all relevant NN approaches applied for CT image reconstruction, bone segmentation, and surgical planning. After full text screening, 76 publications were identified: 32 focusing on CT image reconstruction, 33 focusing on bone segmentation and 11 focusing on surgical planning. Generally, convolutional NNs were most widely used in the identified studies, although the multilayer perceptron was most commonly applied in surgical planning tasks. Moreover, the drawbacks of current approaches and promising research avenues are discussed.

Physics-assisted generative adversarial network for X-ray tomography

X-ray tomography is capable of imaging the interior of objects in three dimensions non-invasively, with applications in biomedical imaging, materials science, electronic inspection, and other fields. The reconstruction process can be an ill-conditioned inverse problem, requiring regularization to obtain satisfactory results. Recently, deep learning has been adopted for tomographic reconstruction. Unlike iterative algorithms which require a distribution that is known a priori, deep reconstruction networks can learn a prior distribution through sampling the training distributions. In this work, we develop a Physics-assisted Generative Adversarial Network (PGAN), a two-step algorithm for tomographic reconstruction. In contrast to previous efforts, our PGAN utilizes maximum-likelihood estimates derived from the measurements to regularize the reconstruction with both known physics and the learned prior. Compared with methods with less physics assisting in training, PGAN can reduce the photon requirement with limited projection angles to achieve a given error rate. The advantages of using a physics-assisted learned prior in X-ray tomography may further enable low-photon nanoscale imaging.

Deep Learning-Based Image Reconstruction for Different Medical Imaging Modalities

Image reconstruction in magnetic resonance imaging (MRI) and computed tomography (CT) is a mathematical process that generates images at many different angles around the patient. Image reconstruction has a fundamental impact on image quality. In recent years, the literature has focused on deep learning and its applications in medical imaging, particularly image reconstruction. Due to the performance of deep learning models in a wide variety of vision applications, a considerable amount of work has recently been carried out using image reconstruction in medical images. MRI and CT appear as the ultimate scientifically appropriate imaging mode for identifying and diagnosing different diseases in this ascension age of technology. This study demonstrates a number of deep learning image reconstruction approaches and a comprehensive review of the most widely used different databases. We also give the challenges and promising future directions for medical image reconstruction.

Multi-domain integrative Swin transformer network for sparse-view tomographic reconstruction

Decreasing projection views to a lower X-ray radiation dose usually leads to severe streak artifacts. To improve image quality from sparse-view data, a multi-domain integrative Swin transformer network (MIST-net) was developed and is reported in this article. First, MIST-net incorporated lavish domain features from data, residual data, image, and residual image using flexible network architectures, where a residual data and residual image sub-network was considered as a data consistency module to eliminate interpolation and reconstruction errors. Second, a trainable edge enhancement filter was incorporated to detect and protect image edges. Third, a high-quality reconstruction Swin transformer (i.e., Recformer) was designed to capture image global features. The experimental results on numerical and real cardiac clinical datasets with 48 views demonstrated that our proposed MIST-net provided better image quality with more small features and sharp edges than other competitors.

The use of deep learning methods in low-dose computed tomography image reconstruction: a systematic review

Conventional reconstruction techniques, such as filtered back projection (FBP) and iterative reconstruction (IR), which have been utilised widely in the image reconstruction process of computed tomography (CT) are not suitable in the case of low-dose CT applications, because of the unsatisfying quality of the reconstructed image and inefficient reconstruction time. Therefore, as the demand for CT radiation dose reduction continues to increase, the use of artificial intelligence (AI) in image reconstruction has become a trend that attracts more and more attention. This systematic review examined various deep learning methods to determine their characteristics, availability, intended use and expected outputs concerning low-dose CT image reconstruction. Utilising the methodology of Kitchenham and Charter, we performed a systematic search of the literature from 2016 to 2021 in Springer, Science Direct, arXiv, PubMed, ACM, IEEE, and Scopus. This review showed that algorithms using deep learning technology are superior to traditional IR methods in noise suppression, artifact reduction and structure preservation, in terms of improving the image quality of low-dose reconstructed images. In conclusion, we provided an overview of the use of deep learning approaches in low-dose CT image reconstruction together with their benefits, limitations, and opportunities for improvement.

A Limited-View CT Reconstruction Framework Based on Hybrid Domains and Spatial Correlation

Limited-view Computed Tomography (CT) can be used to efficaciously reduce radiation dose in clinical diagnosis, it is also adopted when encountering inevitable mechanical and physical limitation in industrial inspection. Nevertheless, limited-view CT leads to severe artifacts in its imaging, which turns out to be a major issue in the low dose protocol. Thus, how to exploit the limited prior information to obtain high-quality CT images becomes a crucial issue. We notice that almost all existing methods solely focus on a single CT image while neglecting the solid fact that, the scanned objects are always highly spatially correlated. Consequently, there lies bountiful spatial information between these acquired consecutive CT images, which is still largely left to be exploited. In this paper, we propose a novel hybrid-domain structure composed of fully convolutional networks that groundbreakingly explores the three-dimensional neighborhood and works in a “coarse-to-fine” manner. We first conduct data completion in the Radon domain, and transform the obtained full-view Radon data into images through FBP. Subsequently, we employ the spatial correlation between continuous CT images to productively restore them and then refine the image texture to finally receive the ideal high-quality CT images, achieving PSNR of 40.209 and SSIM of 0.943. Besides, unlike other current limited-view CT reconstruction methods, we adopt FBP (and implement it on GPUs) instead of SART-TV to significantly accelerate the overall procedure and realize it in an end-to-end manner.

A novel simulation-driven reconstruction approach for X-ray computed tomography

PURPOSE

Radiation dose reduction is critical to the success of x-ray computed tomography (CT). Many advanced reconstruction techniques have been developed over the years to combat noise resulting from the low-dose CT scans. These algorithms rely on accurate local estimation of the image noise to determine reconstruction parameters or to select inferencing models. Because of difficulties in the noise estimation for heterogeneous objects, the performance of many algorithms is inconsistent and suboptimal. In this paper, we propose a novel approach to overcome such shortcoming.

METHOD

By injecting appropriate amount of noise in the CT raw data, a computer simulation approach is capable of accurately estimating the local statistics of the raw data and the local noise in the reconstructed images. This information is then used to guide the noise reduction process during the reconstruction. As an initial implementation, a scaling map is generated based on the noise predicted from the simulation and the noise estimated from existing reconstruction algorithms. Images generated with existing algorithms are subsequently modified based on the scaling map. In this study, both iterative reconstruction (IR) and deep learning image reconstruction (DLIR) algorithms are evaluated.

RESULTS

Phantom experiments were conducted to evaluate the performance of the simulation-based noise estimation in terms of the standard deviation and noise power spectrum (NPS). Quantitative results have demonstrated that the noise measured from the original image matches well with the noise estimated from the simulation. Clinical datasets were utilized to further confirm the accuracy of the proposed approach under more challenging conditions. To validate the performance of the proposed reconstruction approach, clinical scans were used. Performance comparison was carried out qualitatively and quantitatively. Two existing advanced reconstruction techniques, IR and DLIR, were evaluated against the proposed approach. Results have shown that the proposed approach outperforms existing IR and DLIR algorithms in terms of noise suppression and, equally importantly, noise uniformity across the entire imaging volume. Visual assessment of the images also reveals that the proposed approach does not endure noise texture issues facing some of the existing reconstruction algorithms today.

CONCLUSION

Phantom and clinical results have demonstrated superior performance of the proposed approach with regard to noise reduction as well as noise homogeneity. Visual inspection of the noise texture further confirms the clinical utility of the proposed approach. Future enhancements on the current implementation are explored regarding image quality and computational efficiency. Because of the limited scope of this paper, detailed investigation on these enhancement features will be covered in a separate report. This article is protected by copyright. All rights reserved.

Computerized Tomography Image Features under the Reconstruction Algorithm in the Evaluation of the Effect of Ropivacaine Combined with Dexamethasone and Dexmedetomidine on Assisted Thoracoscopic Lobectomy

This research was aimed to study CT image features based on the backprojection filtering reconstruction algorithm and evaluate the effect of ropivacaine combined with dexamethasone and dexmedetomidine on assisted thoracoscopic lobectomy to provide reference for clinical diagnosis. A total of 110 patients undergoing laparoscopic resection were selected as the study subjects. Anesthesia induction and nerve block were performed with ropivacaine combined with dexamethasone and dexmedetomidine before surgery, and chest CT scan was performed. The backprojection image reconstruction algorithm was constructed and applied to patient CT images for reconstruction processing. The results showed that when the overlapping step size was 16 and the block size was 32 × 32, the running time of the algorithm was the shortest. The resolution and sharpness of reconstructed images were better than the Fourier transform analytical method and iterative reconstruction algorithm. The detection rates of lung nodules smaller than 6 mm and 6–30 mm (92.35% and 95.44%) were significantly higher than those of the Fourier transform analytical method and iterative reconstruction algorithm (90.98% and 87.53%; 88.32% and 90.87%) (P < 0.05). After anesthesia induction and lobectomy with ropivacaine combined with dexamethasone and dexmedetomidine, the visual analogue scale (VAS) decreased with postoperative time. The VAS score decreased to a lower level (1.76 ± 0.54) after five days. In summary, ropivacaine combined with dexamethasone and dexmedetomidine had better sedation and analgesia effects in patients with thoracoscopic lobectomy. CT images based on backprojection reconstruction algorithm had a high recognition accuracy for lung lesions.

Mutual Information-Based Non-Local Total Variation Denoiser for Low-Dose Cone-Beam Computed Tomography

Conventional non-local total variation (NLTV) approaches use the weight of a non-local means (NLM) filter, which degrades performance in low-dose cone-beam computed tomography (CBCT) images generated with a low milliampere-seconds (mAs) parameter value because a local patch used to determine the pixel weights comprises noisy-damaged pixels that reduce the similarity between corresponding patches. In this paper, we propose a novel type of NLTV based on a combination of mutual information (MI): MI-NLTV. It is based on a statistical measure for a similarity calculation between the corresponding bins of non-local patches vs. a reference patch. The weight is determined in terms of a statistical measure comprising the MI value between corresponding non-local patches and the reference-patch entropy. The MI-NLTV denoising process is applied to CBCT images generated by the analytical reconstruction algorithm using a ray-driven backprojector (RDB). The MI-NLTV objective function is minimized based on the steepest gradient descent optimization to augment the difference between a real structure and noise, cleaning noisy pixels without significant loss of the fine structure and details that remain in the reconstructed images. The proposed method was evaluated using patient data and actual phantom measurement data acquired with lower mAs. The results show that integrating the RDB further enhances the MI-NLTV denoising-based analytical reconstruction algorithm to achieve a higher CBCT image quality when compared with those generated by NLTV denoising-based approach, with an average of 15.97% higher contrast-to-noise ratio, 2.67% lower root mean square error, 0.12% lower spatial non-uniformity, 1.14% higher correlation, and an average of 18.11% higher detectability index. These quantitative results indicate that the incorporation of MI makes the NLTV more stable and robust than the conventional NLM filter for low-dose CBCT imaging. In addition, achieving clinically acceptable CBCT image quality despite low-mAs projection acquisition can reduce the burden on common online CBCT imaging, improving patient safety throughout the course of radiotherapy.

A Survey of Soft Computing Approaches in Biomedical Imaging

Medical imaging is an essential technique for the diagnosis and treatment of diseases in modern clinics. Soft computing plays a major role in the recent advances in medical imaging. It handles uncertainties and improves the qualities of an image. Until now, various soft computing approaches have been proposed for medical applications. This paper discusses various medical imaging modalities and presents a short review of soft computing approaches such as fuzzy logic, artificial neural network, genetic algorithm, machine learning, and deep learning. We also studied and compared each approach used for other imaging modalities based on the certain parameter used for the system evaluation. Finally, based on comparative analysis, the possible research strategies for further development are proposed. As far as we know, no previous work examined this issue.

Confronting the threat of SARS-CoV-2: Realities, challenges and therapeutic strategies (Review)

The novel coronavirus (SARS-CoV-2) appeared in2019 in Wuhan, China, and rapidly developed into a global pandemic. The disease has affected not only health care systems and economies worldwide but has also changed the lifestyles and habits of the majority of the world's population. Among the potential targets for SARS-CoV-2 therapy, the viral spike glycoprotein has been studied most intensely, due to its key role in mediating viral entry into target cells and inducing a protective antibody response in infected individuals. In the present manuscript the molecular mechanisms that are responsible for SARS-CoV-2 infection are described and a progress report on the status of SARS-CoV-2 research is provided. A brief review of the clinical symptoms of the condition and current diagnostic methods and treatment plans for SARS-CoV-2 are also presented and the progress of preclinical research into medical intervention against SARS-CoV-2 infection are discussed.