Prior Image Research Articles

A prior image constraint robust principal component analysis reconstruction method for sparse segmental multi-energy computed tomography.

Multi-energy computed tomography (MECT) is a promising technique in medical imaging, especially for quantitative imaging. However, high technical requirements and system costs barrier its step into clinical practice. We propose a novel sparse segmental MECT (SSMECT) scheme and corresponding reconstruction method, which is a cost-efficient way to realize MECT on a conventional single-source CT. For the data acquisition, the X-ray source is controlled to maintain an energy within a segmental arc, and then switch alternately to another energy level. This scan only needs to switch tube voltage a few times to acquire multi-energy data, but leads to sparse-view and limited-angle issues in image reconstruction. To solve this problem, we propose a prior image constraint robust principal component analysis (PIC-RPCA) reconstruction method, which introduces structural similarity and spectral correlation into the reconstruction. A numerical simulation and a real phantom experiment were conducted to demonstrate the efficacy and robustness of the scan scheme and reconstruction method. The results showed that our proposed reconstruction method could have achieved better multi-energy images than other competing methods both quantitatively and qualitatively. Our proposed SSMECT scan with PIC-RPCA reconstruction method could lower kVp switching frequency while achieving satisfactory reconstruction accuracy and image quality.

Hybrid Regularization Algorithm for Efficient Image Deblurring

Image deblurring is a challenging illposed problem with widespread applications. Most existing deblurring methods make use of image priors or priors on the PSF to achieve accurate results. The performance of these methods depends on various factors such as the presence of well-lit conditions in the case of dark image priors and in case of statistical image priors the assumption the image follows a certain distribution might not be fully accurate. This holds for statistical priors used on the blur kernel as well. The aim of this paper is to propose a novel image deblurring method which can be readily extended to various applications such that it effectively deblurs the image irrespective of the various factors affecting its capture. A hybrid regularization method is proposed which uses a TV regularization framework with varying sparsity inducing priors. The edges of the image are accurately recovered due to the TV regularization. The sparsity prior is implemented through a dictionary such that varying weights of sparsity is induced based on the different image regions. This helps in smoothing the unwanted artifacts generated due to blur in the uniform regions of the image.

Noise2Context: Context-assisted learning 3D thin-layer for low-dose CT.

Computed tomography (CT) has played a vital role in medical diagnosis, assessment, and therapy planning, etc. In clinical practice, concerns about the increase of x-ray radiation exposure attract more and more attention. To lower the x-ray radiation, low-dose CT (LDCT) has been widely adopted in certain scenarios, while it will induce the degradation of CT image quality. In this paper, we proposed a deep learning-based method that can train denoising neural networks without any clean data. In this work, for 3D thin-slice LDCT scanning, we first drive an unsupervised loss function which was equivalent to a supervised loss function with paired noisy and clean samples when the noise in the different slices from a single scan was uncorrelated and zero-mean. Then, we trained the denoising neural network to map one noise LDCT image to its two adjacent LDCT images in a single 3D thin-layer LDCT scanning, simultaneously. In essence, with some latent assumptions, we proposed an unsupervised loss function to train the denoising neural network in an unsupervised manner, which integrated the similarity between adjacent CT slices in 3D thin-layer LDCT. Further experiments on Mayo LDCT dataset and a realistic pig head were carried out. In the experiments using Mayo LDCT dataset, our unsupervised method can obtain performance comparable to that of the supervised baseline. With the realistic pig head, our method can achieve optimal performance at different noise levels as compared to all the other methods that demonstrated the superiority and robustness of the proposed Noise2Context. In this work, we present a generalizable LDCT image denoising method without any clean data. As a result, our method not only gets rid of the complex artificial image priors but also amounts of paired high-quality training datasets.

Low-dose cerebral perfusion CT image restoration using prior image constrained diffusion tensor

We propose an efficient method to reduce the noise in low-dose cerebral perfusion CT images using prior image constrained diffusion tensor to reduce the radiation dose in brain CT examination. By utilizing the redundant information in cerebral perfusion CT images, we embedded the complementary structure information in prior images into lowdose cerebral perfusion CT image restoration process to suppress the image noise and artifacts.We first calculated the diffusion tensor for the low-dose cerebral perfusion CT image and prior image separately and then constructed a prior image constrained diffusion tensor (PICDT) to incorporate the structure information from the prior image into low-dose image restoration process. In experiments with the Shepp-Logan phantom, the SSIM value of CBF map obtained by the proposed algorithm was increased by 63% as compared with that of the FBP algorithm.In analysis of the clinical dataset, the SSIM value of CBF map obtained by the proposed algorithm was increased by 45% as compared with that of FBP algorithm. The proposed method can effectively reduce noises and artifacts of low-dose cerebral perfusion CT images while maintaining the structural details to obtain accurate cerebral hemodynamic maps.

MRI-aided kernel PET image reconstruction method based on texture features

We investigate the reconstruction of low-count positron emission tomography (PET) projection, which is an important, but challenging, task. Using the texture feature extraction method of radiomics, i.e. the gray-level co-occurrence matrix (GLCM), texture features can be extracted from magnetic resonance imaging images with high-spatial resolution. In this work, we propose a kernel reconstruction method combining autocorrelation texture features derived from the GLCM. The new kernel function includes the correlations of both the intensity and texture features from the prior image. By regarding the GLCM as a discrete approximation of a probability density function, the asymptotically gray-level-invariant autocorrelation texture feature is generated, which can maintain the accuracy of texture features extracted from small image regions by reducing the number of quantized image gray levels. A computer simulation shows that the proposed method can effectively reduce the noise in the reconstructed image compared to the maximum likelihood expectation maximum method and improve the image quality and tumor region accuracy compared to the original kernel method for low-count PET reconstruction. A simulation study on clinical patient images also shows that the proposed method can improve the whole image quality and that the reconstruction of a high-uptake lesion is more accurate than that achieved by the original kernel method.

Quantitative Analysis of Aspergillus nidulans Growth Rate using Live Microscopy and Open-Source Software.

It is well established that colony growth of filamentous fungi, mostly dependent on changes in hyphae/mycelia apical growth rate, is macroscopically estimated on solidified media by comparing colony size. However, to quantitatively measure the growth rate of genetically different fungal strains or strains under different environmental/growth conditions (pH, temperature, carbon and nitrogen sources, antibiotics, etc.) is challenging. Thus, the pursuit of complementary approaches to quantify growth kinetics becomes mandatory in order to better understand fungal cell growth. Furthermore, it is well-known that filamentous fungi, including Aspergillus spp., have distinct modes of growth and differentiation under sub-aerial conditions on solid media or submerged cultures. Here, we detail a quantitative microscopic method for analyzing growth kinetics ofthe model fungus Aspergillus nidulans, using live imaging in both submerged cultures and solid media. We capture images, analyze, and quantify growth rates of different fungal strains in a reproducible and reliable manner using an open source, free software for bio-images (e.g., Fiji), in a way that does not require any prior image analysis expertise from the user.

Populational and individual information based PET image denoising using conditional unsupervised learning

Our study aims to improve the signal-to-noise ratio of positron emission tomography (PET) imaging using conditional unsupervised learning. The proposed method does not require low- and high-quality pairs for network training which can be easily applied to existing PET/computed tomography (CT) and PET/magnetic resonance (MR) datasets. This method consists of two steps: populational training and individual fine-tuning. As for populational training, a network was first pre-trained by a group of patients’ noisy PET images and the corresponding anatomical prior images from CT or MR. As for individual fine-tuning, a new network with initial parameters inherited from the pre-trained network was fine-tuned by the test patient's noisy PET image and the corresponding anatomical prior image. Only the last few layers were fine-tuned to take advantage of the populational information and the pre-training efforts. Both networks shared the same structure and took the CT or MR images as the network input so that the network output was conditioned on the patient’s anatomic prior information. The noisy PET images were used as the training and fine-tuning labels. The proposed method was evaluated on a 68Ga-PPRGD2 PET/CT dataset and a 18F-FDG PET/MR dataset. For the PET/CT dataset, with the original noisy PET image as the baseline, the proposed method has a significantly higher contrast-to noise ratio (CNR) improvement (71.85% ± 27.05%) than Gaussian (12.66% ± 6.19%, P = 0.002), nonlocal mean method (22.60% ± 13.11%, P = 0.002) and conditional deep image prior method (52.94% ± 21.79%, P = 0.0039). For the PET/MR dataset, compared to Gaussian (18.73% ± 9.98%, P < 0.0001), NLM (26.01% ± 19.40%, P < 0.0001) and CDIP (47.48% ± 25.36%, P < 0.0001), the CNR improvement ratio of the proposed method (58.07% ± 28.45%) is the highest. In addition, the denoised images using both datasets also showed that the proposed method can accurately restore tumor structures while also smoothing out the noise.

Remote Sensing-Guided Spatial Sampling Strategy over Heterogeneous Surface Ground for Validation of Vegetation Indices Products with Medium and High Spatial Resolution

Remote sensing (RS)-derived vegetation indices (VIs) with medium and high spatial resolution have emerged as a promising dataset for fine-scale ecosystem modeling and agricultural monitoring at local or global scales. Before they can be used as reliable inputs for other research, conducting in situ measurements for validation is very critical. However, the spatial heterogeneity due to the diversity of land cover and its spatial organization in the landscape increases the uncertainty of validation, so design of optimal sampling is an important basis for the reliability of the validation. In this paper, we propose an integrative stratified sampling strategy (INTEG-STRAT) based on normalized difference vegetation index (NDVI) data as prior knowledge. The basic idea is to realize a sampling optimization by determining the optimal combination of the spatial sampling method (e.g., simple random sampling (SRS), spatial system sampling (SYS), stratified sampling, generalized random tessellation stratified (GRTS), balanced acceptance sampling (BAS)) and spatial stratification scheme with an objective rule. The objective rule in this paper is to minimize the root mean square error (RMSE) of 10-fold cross validation between estimated values (sample are not included) and the corresponding values on prior knowledge. Relative precision, correlation coefficient, and RMSE are used to compare the effectiveness of the proposed sampling strategy with each sampling method without considering sampling optimization. After comparing, we find that the INTEG-STRAT requires fewer samples to become stable and has higher accuracy. At site 1, when the correlation coefficient between NDVI image and the simulated NDVI surface reached 80%, INTEG-STRAT needed only 70 sampling points while other methods require more sampling points. At the same time, INTEG-STRAT strategy has a smaller RMSE between the estimated values and the corresponding values on prior knowledge image. In general, INTEG-STRAT is an effective method in the selection of representative samples to support the validation of vegetation indices products with medium and high spatial resolution.

Cygnus A jointly calibrated and imaged via non-convex optimization from VLA data

ABSTRACT Radio interferometric (RI) data are noisy undersampled spatial Fourier components of the unknown radio sky affected by direction-dependent antenna gains. Failure to model these antenna gains accurately results in a radio sky estimate with limited fidelity and resolution. The RI inverse problem has been recently addressed via a joint calibration and imaging approach that consists in solving a non-convex minimization task, involving suitable priors for the direction-dependent effects (DDEs), namely temporal and spatial smoothness, and sparsity for the unknown radio map via an ℓ1-norm prior, in the context of realistic RI simulations. Building on these developments, we propose to promote sparsity of the radio map via a log-sum prior, enforcing sparsity more strongly than the ℓ1 norm. The resulting minimization task is addressed via a sequence of non-convex minimization tasks composed of re-weighted ℓ1 image priors, which are solved approximately. We demonstrate the efficiency of the approach on RI observations of the celebrated radio galaxy Cygnus A obtained with the Karl G. Jansky Very Large Array at the frequency bands X, C, and S . More precisely, we showcase that the approach enhances data fidelity significantly while achieving high-resolution high-dynamic range radio maps, confirming the suitability of the priors considered for the unknown DDEs and radio image. As a clear qualitative indication of the high fidelity achieved by the data and the proposed approach, we report the detection of three background sources in the vicinity of Cyg A, at S band.

Interpolating missing land cover data using stochastic spatial random forests for improved change detection

AbstractForest cover requires large scale and frequent monitoring as an indicator of biodiversity and progress towards United Nations and World Bank Sustainable Development Goal 15. Measuring change in forest cover over time is an essential task in order to track and preserve quality habitats for species around the world. Due to the prohibitive expense and impracticality of mass field data collection to monitor forest cover at regular intervals, satellite images are a key data source for monitoring forest cover globally. A challenge of working with satellite images is missing data due to clouds. Existing methods for interpolating the missing data based on past images, such as compositing, are effective for stable land cover but can be inaccurate for dynamic and substantially changing landscapes. Here we present an adaptation of our recent stochastic spatial random forest (SS‐RF) method, which combines observed data from a prior image and modelled estimates of the current image to produce interpolated land cover values and associated probabilities of those values. Results show our SS‐RF method accurately detected simulated land cover change under both clear felling (0.83 average overall accuracy) and tree thinning (0.85 average overall accuracy). Our method detected forest cover change substantially more accurately than compositing, offering 39% and 12% increases in average overall accuracy for clear felling and tree thinning simulations respectively. However, when natural fluctuation occurs and there is minimal change in land cover, compositing has equivalent or more accurate performance than our method. Overall we find that our SS‐RF method produces accurate estimates under a range of simulated forest clearing scenarios and has a more accurate and robust performance than compositing when modelling noticeably changing landscapes.

Adaptive Sparsity through Hybrid Regularization for Effective Image Deblurring

Objectives: To develop image deblurring algorithm to effectively recover the original image from the captured blurry and noisy image. Methods: A novel image deblurring algorithm using adaptive priors is proposed. The adaptive priors vary the sparsity induced based on factors such as the type of blur affecting the image, image region, statistical parameters regarding the blur kernel and/or statistical predictions on the blur kernel. The proposed algorithm using adaptive priors improves the quality of the deblurred result, when compared with various recent image deblurring algorithms. Finding: the work contains a case study with regard to certain standard parameters. It is observed that proposed method is better in terms of frequency response, Peak signal-tonoise ratio (PSNR) and Structural Similarity Index (SSIM) values in comparison with other priors. Novelty: the proposed priors lead to the most effective results for image deblurring using the Bayesian framework. The proposed method improves the performance by 30% in PSNR and 45% in SSIM values in dB with uniform kernel size 12 * 12 and improves performance by 30 % in PSNR and 32% SSIM with a standard deviation of 3.5. The proposed method enhances the frequency response of the real-time image restoration process. Applications: Some important applications include the restoration of medical images such as MRI images, CT images where the intensity of these radiations is maintained to avoid damage to human organs. Remote sensing images captured through drones at a specified time cannot be retaken and hence restoring such images is very important. Similarly restoring images from CCTV footage, astronomical images. Keywords: Blind Image Deblurring; Maximum a Posteriori Estimation; Image priors; Regularization; Structural Similarity Index (SSIM) and Peak signal-to-noise ratio (PSNR)

Physics-based iterative reconstruction for dual-source and flying focal spot computed tomography.

For single-source helical Computed Tomography (CT), both Filtered-Back Projection (FBP) and statistical iterative reconstruction have been investigated. However, for dual-source CT with flying focal spot (DS-FFS CT), a statistical iterative reconstruction that accurately models the scanner geometry and acquisition physics remains unknown to researchers. Therefore, our purpose is to present a novel physics-based iterative reconstruction method for DS-FFS CT and assess its image quality. Our algorithm uses precise physics models to reconstruct from the native cone-beam geometry and interleaved dual-source helical trajectory of a DS-FFS CT. To do so, we construct a noise physics model to represent data acquisition noise and a prior image model to represent image noise and texture. In addition, we design forward system models to compute the locations of deflected focal spots, the dimension, and sensitivity of voxels and detector units, as well as the length of intersection between x-rays and voxels. The forward system models further represent the coordinated movement between the dual sources by computing their x-ray coverage gaps and overlaps at an arbitrary helical pitch. With the above models, we reconstruct images by an advanced Consensus Equilibrium (CE) numerical method to compute the maximum a posteriori estimate to a joint optimization problem that simultaneously fits all models. We compared our reconstruction with Siemens ADMIRE, which is the clinical standard hybrid iterative reconstruction (IR) method for DS-FFS CT, in terms of spatial resolution, noise profile, and image artifacts through both phantoms and clinical scan datasets. Experiments show that our reconstruction has a higher spatial resolution, with a Task-Based Modulation Transfer Function (MTFtask ) consistently higher than the clinical standard hybrid IR. In addition, our reconstruction shows a reduced magnitude of image undersampling artifacts than the clinical standard. By modeling a precise geometry and avoiding data rebinning or interpolation, our physics-based reconstruction achieves a higher spatial resolution and fewer image artifacts with smaller magnitude than the clinical standard hybrid IR.

Color-guided optimization model with reliable self-structure priors for depth map restoration

Depth maps captured by Kinect or time of flight (ToF) cameras have an active role in many visual applications. However, a brutal truth is that these depth maps are often contaminated with compound noise, which includes intrinsic noise and missing pixels. In addition, depth maps captured with ToF-based cameras are low in resolution. As these depth maps carry rich and critical information about 3D space, high quality post-processing is crucial for supporting subsequent visual applications. Previous works were proposed via the guiding of the registered color image and bicubic interpolation as an initialization for the up-sampling task, where challenges arose from texture coping and blurry depth discontinuities. Motivated by these challenges, in this paper, we propose a new optimization model depending on the relative structures of both depth and color images for both depth map filtering and up-sampling tasks. In our general model, two self-structure priors for depth and color images are constructed individually and used for the two tasks. For overcoming the texture coping problem, the color-based and depth-based priors are used near the depth edges and at the homogeneous regions respectively. To this end, we further propose a confidence map at every task for managing where every prior is used. Experimental results on both simulated and real datasets for Kinect and ToF cameras demonstrate that the proposed method has a superior performance than benchmarks.

Preventing face morphing attacks by using legacy face images

Countries allow citizens to upload a face image or provide printed copies to authorities to issue their passport. This allows prior image manipulation with criminal intent. A composite image can be created by blending the images of two individuals before submitting the composite image to the authorities. Depending on several factors, the submitted morphed face image can fool the issuing officer to issue a legitimate document. The document can then be successfully used by either contributor to attack the automatic Face Recognition Systems (FRS) operating, for example, at Automatic Border Control (ABC) airport gates. This is known as a Morphing Attack (MA), an identity sharing scheme with serious consequences. Here, the security vulnerabilities due to MAs are identified and analysed, and an additional security measure that allows mitigating the risk or preventing MAs in certain scenarios is proposed. The measure introduces more comparisons by keeping the old passport or ID card image in the chip, in passport renewal applications or first time passport applications, respectively. This approach is implemented with two FRSs on a challenging dataset and the dramatic decrease in the vulnerability is shown. Finally, their performance is compared with a state-of-the-art MA detection algorithm on the same dataset.

Iterative reconstruction for low-dose cerebral perfusion computed tomography using prior image induced diffusion tensor

Cerebral perfusion computed tomography (CPCT) can depict the functional status of cerebral circulation at the tissue level; hence, it has been increasingly used to diagnose patients with cerebrovascular disease. However, there is a significant concern that CPCT scanning protocol could expose patients to excessive radiation doses. Although reducing the x-ray tube current when acquiring CPCT projection data is an effective method for reducing radiation dose, this technique usually results in degraded image quality. To enhance the image quality of low-dose CPCT, we present a prior image induced diffusion tensor (PIDT) for statistical iterative reconstruction, based on the penalized weighted least-squares (PWLS) criterion, which we referred to as PWLS-PIDT, for simplicity. Specifically, PIDT utilizes the geometric features of pre-contrast scanned high-quality CT image as a structure prior for PWLS reconstruction; therefore, the low-dose CPCT images are enhanced while preserving important features in the target image. An effective alternating minimization algorithm is developed to solve the associated objective function in the PWLS-PIDT reconstruction. We conduct qualitative and quantitative studies to evaluate the PWLS-PIDT reconstruction with a digital brain perfusion phantom and patient data. With this method, the noise in the reconstructed CPCT images is more substantially reduced than that of other competing methods, without sacrificing structural details significantly. Furthermore, the CPCT sequential images reconstructed via the PWLS-PIDT method can derive more accurate hemodynamic parameter maps than those of other competing methods.

Photon-counting normalized metal artifact reduction (NMAR) in diagnostic CT.

Metal artifacts can drastically reduce the diagnostic value of computed tomography (CT) images. Even the state-of-the-art algorithms cannot remove them completely. Photon-counting CT inherently provides spectral information, similar to dual-energy CT. Many applications, such as material decomposition, are not possible when metal artifacts are present. Our aim is to develop a prior-based metal artifact reduction specifically for photon-counting CT that can correct each bin image individually or in their combinations. Photon-counting CT sorts incoming photons into several energy bins, producing bin and threshold images containing spectral information. We use this spectral information to obtain a better prior image for the state-of-the-art metal artifact reduction algorithm FSNMAR. First, we apply a non-linear transformation to the bin images to obtain bone-emphasized images. Subsequently, we forward-project the bin images and bone-emphasized images and multiply the resulting sinograms with each other element-wise to mimic beam hardening effects. These sinograms are reconstructed and linearly combined to produce an artifact-reduced image. The coefficients of this linear combination are automatically determined by minimizing a threshold-based cost function in the image domain. After thresholding, we obtain the prior image for FSNMAR, which is applied to the individual bin images and the lowest threshold image. We test our photon-counting normalized metal artifact reduction (PCNMAR) on forensic CT data and compare it to conventional FSNMAR, where the prior is generated via linear sinogram inpainting. For numerical analysis, we compute both the standard deviation in an ROI with metal artifacts and the CNR of soft tissue and fat. PCNMAR can effectively reduce metal artifacts without sacrificing the overall image quality. Compared to FSNMAR, our method produces fewer secondary artifacts and is more consistent with the measurements. Areas that contain metal, air, and soft tissue are more accurate in PCNMAR. In some cases, the standard deviation in the artifact ROI is reduced by more than 50% relative to FSNMAR, while the CNR values are similar. If extreme artifacts are present, PCNMAR is unable to outperform FSNMAR. Using either two, four, or only the highest energy bin to produce the prior image yielded comparable results. PCNMAR is an effective method of reducing metal artifacts in photon-counting CT. The spectral information available in photon-counting CT is highly beneficial for metal artifact reduction, especially the high-energy bin, which inherently contains fewer artifacts. While scanning with four instead of two bins does not provide a better artifact reduction, it allows for more freedom in the selection of energy thresholds.

Holographic optical field recovery using a regularized untrained deep decoder network

Image reconstruction using minimal measured information has been a long-standing open problem in many computational imaging approaches, in particular in-line holography. Many solutions are devised based on compressive sensing (CS) techniques with handcrafted image priors or supervised deep neural networks (DNN). However, the limited performance of CS methods due to lack of information about the image priors and the requirement of an enormous amount of per-sample-type training resources for DNNs has posed new challenges over the primary problem. In this study, we propose a single-shot lensless in-line holographic reconstruction method using an untrained deep neural network which is incorporated with a physical image formation algorithm. We demonstrate that by modifying a deep decoder network with simple regularizers, a Gabor hologram can be inversely reconstructed via a minimization process that is constrained by a deep image prior. The outcoming model allows to accurately recover the phase and amplitude images without any training dataset, excess measurements, or specific assumptions about the object’s or the measurement’s characteristics.

Image Denoising Using a Novel Deep Generative Network with Multiple Target Images and Adaptive Termination Condition

Image denoising, a classic ill-posed problem, aims to recover a latent image from a noisy measurement. Over the past few decades, a considerable number of denoising methods have been studied extensively. Among these methods, supervised deep convolutional networks have garnered increasing attention, and their superior performance is attributed to their capability to learn realistic image priors from a large amount of paired noisy and clean images. However, if the image to be denoised is significantly different from the training images, it could lead to inferior results, and the networks may even produce hallucinations by using inappropriate image priors to handle an unseen noisy image. Recently, deep image prior (DIP) was proposed, and it overcame this drawback to some extent. The structure of the DIP generator network is capable of capturing the low-level statistics of a natural image using an unsupervised method with no training images other than the image itself. Compared with a supervised denoising model, the unsupervised DIP is more flexible when processing image content that must be denoised. Nevertheless, the denoising performance of DIP is usually inferior to the current supervised learning-based methods using deep convolutional networks, and it is susceptible to the over-fitting problem. To solve these problems, we propose a novel deep generative network with multiple target images and an adaptive termination condition. Specifically, we utilized mainstream denoising methods to generate two clear target images to be used with the original noisy image, enabling better guidance during the convergence process and improving the convergence speed. Moreover, we adopted the noise level estimation (NLE) technique to set a more reasonable adaptive termination condition, which can effectively solve the problem of over-fitting. Extensive experiments demonstrated that, according to the denoising results, the proposed approach significantly outperforms the original DIP method in tests on different databases. Specifically, the average peak signal-to-noise ratio (PSNR) performance of our proposed method on four databases at different noise levels is increased by 1.90 to 4.86 dB compared to the original DIP method. Moreover, our method achieves superior performance against state-of-the-art methods in terms of popular metrics, which include the structural similarity index (SSIM) and feature similarity index measurement (FSIM). Thus, the proposed method lays a good foundation for subsequent image processing tasks, such as target detection and super-resolution.

CycN-Net: A Convolutional Neural Network Specialized for 4D CBCT Images Refinement.

Four-dimensional cone-beam computed tomography (4D CBCT) has been developed to provide a sequence of phase-resolved reconstructions in image-guided radiation therapy. However, 4D CBCT images are degraded by severe streaking artifacts and noise because the phase-resolved image is an extremely sparse-view CT procedure wherein a few under-sampled projections are used for the reconstruction of each phase. Aiming at improving the overall quality of 4D CBCT images, we proposed two CNN models, named N-Net and CycN-Net, respectively, by fully excavating the inherent property of 4D CBCT. To be specific, the proposed N-Net incorporates the prior image reconstructed from entire projection data based on U-Net to boost the image quality for each phase-resolved image. Based on N-Net, a temporal correlation among the phase-resolved images is also considered by the proposed CycN-Net. Extensive experiments on both XCAT simulation data and real patient 4D CBCT datasets were carried out to verify the feasibility of the proposed CNNs. Both networks can effectively suppress streaking artifacts and noise while restoring the distinct features simultaneously, compared with the existing CNN models and two state-of-the-art iterative algorithms. Moreover, the proposed method is robust in handling complicated tasks of various patient datasets and imaging devices, which implies its excellent generalization ability.

The adaptation and investigation of cone-beam CT reconstruction algorithms for horizontal rotation fixed-gantry scans of rabbits

Fixed-gantry radiation therapy has been proposed as a low-cost alternative to the conventional rotating-gantry radiation therapy, that may help meet the rising global treatment demand. Fixed-gantry systems require gravitational motion compensated reconstruction algorithms to produce cone-beam CT (CBCT) images of sufficient quality for image guidance. The aim of this work was to adapt and investigate five CBCT reconstruction algorithms for fixed-gantry CBCT images. The five algorithms investigated were Feldkamp–Davis–Kress (FDK), prior image constrained compressed sensing (PICCS), gravitational motion compensated FDK (GMCFDK), motion compensated PICCS (MCPICCS) (a novel CBCT reconstruction algorithm) and simultaneous motion estimation and iterative reconstruction (SMEIR). Fixed-gantry and rotating-gantry CBCT scans were acquired of 3 rabbits, with the rotating-gantry scans used as a reference. Projections were sorted into rotation bins, based on the angle of rotation of the rabbit during image acquisition. The algorithms were compared using the structural similarity index measure root mean square error, and reconstruction time. Evaluation of the reconstructed volumes showed that, when compared with the reference rotating-gantry volume, the conventional FDK algorithm did not accurately reconstruct fixed-gantry CBCT scans. Whilst the PICCS reconstruction algorithm reduced some motion artefacts, the motion estimation reconstruction methods (GMCFDK, MCPICCS and SMEIR) were able to greatly reduce the effect of motion artefacts on the reconstructed volumes. This finding was verified quantitatively, with GMCFDK, MCPICCS and SMEIR reconstructions having RMSE 17%–19% lower and SSIM 1% higher than a conventional FDK. However, all motion compensated fixed-gantry CBCT reconstructions had a 56%–61% higher RMSE and 1.5% lower SSIM than FDK reconstructions of conventional rotating-gantry CBCT scans. The results show that motion compensation is required to reduce motion artefacts for fixed-gantry CBCT reconstructions. This paper further demonstrates the feasibility of fixed-gantry CBCT scans, and the ability of CBCT reconstruction algorithms to compensate for motion due to horizontal rotation.