Medical Image Reconstruction Research Articles

An Overview of Artificial Intelligence Applications in Liver and Pancreatic Imaging.

Simple SummaryArtificial intelligence (AI) is gaining more and more attention in radiology. The efficiency of AI-based algorithms to solve specific problems is, in some cases, far superior compared to human-driven approaches. This is particularly evident in some repetitive tasks, such as segmentation, where AI usually outperforms manual approaches. AI may be also used in quantification where it can provide, for example, fast and efficient longitudinal follow up in liver tumour burden. AI, thanks to the association with radiomic and big data, may also suggest a diagnosis. Finally, AI algorithms can also reduce scan time, increase image quality and, in the case of computed tomography, reduce patient dose.Artificial intelligence (AI) is one of the most promising fields of research in medical imaging so far. By means of specific algorithms, it can be used to help radiologists in their routine workflow. There are several papers that describe AI approaches to solve different problems in liver and pancreatic imaging. These problems may be summarized in four different categories: segmentation, quantification, characterization and image quality improvement. Segmentation is usually the first step of successive elaborations. If done manually, it is a time-consuming process. Therefore, the semi-automatic and automatic creation of a liver or a pancreatic mask may save time for other evaluations, such as quantification of various parameters, from organs volume to their textural features. The alterations of normal liver and pancreas structure may give a clue to the presence of a diffuse or focal pathology. AI can be trained to recognize these alterations and propose a diagnosis, which may then be confirmed or not by radiologists. Finally, AI may be applied in medical image reconstruction in order to increase image quality, decrease dose administration (referring to computed tomography) and reduce scan times. In this article, we report the state of the art of AI applications in these four main categories.

Generative Adversarial Networks in Medical Image Processing.

The emergence of generative adversarial networks (GANs) has provided new technology and framework for the application of medical images. Specifically, a GAN requires little to no labeled data to obtain high-quality data that can be generated through competition between the generator and discriminator networks. Therefore, GANs are rapidly proving to be a state-of-the-art foundation, achieving enhanced performances in various medical applications. In this article, we introduce the principles of GANs and their various variants, deep convolutional GAN, conditional GAN, Wasserstein GAN, Info-GAN, boundary equilibrium GAN, and cycle-GAN. All various GANs have found success in medical imaging tasks, including medical image enhancement, segmentation, classification, reconstruction, and synthesis. Furthermore, we summarize the data processing methods and evaluation indicators. Finally, we note the limitations of existing methods and the existing challenges that need to be addressed in this field. Although GANs are in the initial stage of development in medical image processing, it will have a great prospect in the future.

Design of Cascaded CNN for Medical Image Super Resolution

The images captured through a camera usually belong to over or under exposed conditions. The reason may be inappropriate lighting conditions or camera resolution. Hence, it is of utmost importance to have a few enhancement techniques that could make these artefacts look better. Hence, the primary objective pertaining to the adjustment and enhancement techniques is to enhance the characteristics of an image. The initial numeric values related to an image get distorted when an image is enhanced. Therefore, enhancement techniques should be designed in such a way that the image quality isn’t compromised. This research work is focused on proposed a network design for deep convolution neural networks for application of super resolution techniques. To improve the complexity of existing techniques this work is intended towards network designs, different filter size and CNN architecture. The CNN model is most effective model for detection and segmentation in image. This model will improve the efficiency of medical image reconstruction from LR to HR. The proposed model showed its efficiency not only PET medical images but also on retinal database and achieved advance results as compared to existing works.

SpiNet: A deep neural network for Schatten p-norm regularized medical image reconstruction.

To propose a generic deep learning based medical image reconstruction model (named as SpiNet) that can enforce any Schatten p-norm regularization with 0<p≤2, where the p can be learnt (or fixed) based on the problem at hand. Model-based deep learning architecture for solving inverse problems consists of two parts, a deep learning based denoiser and an iterative data consistency solver. The former has either L2 norm or L1 norm enforced on it, which are convex and can be easily minimized. This work proposes a method to enforce any p norm on the noise prior where 0<p≤2. This is achieved by using Majorization-Minimization algorithm, which upper bounds the cost function with a convex function, thus can be easily minimized. The proposed SpiNet has the capability to work for a fixed p or it can learn p based on the data. The network was tested for solving the inverse problem of reconstructing magnetic resonance (MR) images from undersampled k space data and the results were compared with a popular model-based deep learning architecture MoDL which enforces L2 norm along with other compressive sensing-based algorithms. This comparison between MoDL and proposed SpiNet was performed for undersampling rates (R) of 2×, 4×, 6×, 8×, 12×, 16×, and 20×. Multiple figures of merit such as PSNR, SSIM, and NRMSE were utilized in this comparison. A two-tailed t test was performed for all undersampling rates and for all metrices for proving the superior performance of proposed SpiNet compared to MoDL. For training and testing, the same dataset that was utilized in MoDL implementation was deployed. The results indicate that for all undersampling rates, the proposed SpiNet shows higher PSNR and SSIM and lower NRMSE than MoDL. However, for low undersampling rates of 2× and 4×, there is no significant difference in performance of proposed SpiNet and MoDL in terms of PSNR and NRMSE. This can be expected as the learnt p value is close to 2 (norm enforced by MoDL). For higher undersampling rates ≥6×, SpiNet significantly outperforms MoDL in all metrices with improvement as high as 4dB in PSNR and 0.5 points in SSIM. As deep learning based medical image reconstruction methods are gaining popularity, the proposed SpiNet provides a generic framework to incorporate Schatten p-norm regularization with 0<p≤2 with an added advantage of providing superior performance compared to its counterparts. The proposed SpiNet also has useful addition of Schatten p-norm value in regularization term being automatically chosen based on the available training data.

Two-step methods and relaxed two-step methods for solving the split equality problem

The inverse problems as a classical mathematical model arise frequently in practical applications. While the split equality problem is an inverse problem arising frequently in signal processing, phase retrieval and medical image reconstruction etc. which is also an important extension of the split feasibility problem. In this paper, motivated and inspired by an idea for solving a variational inequality, we proved that several two-step methods and relaxed two-step methods can be applied to solve the split equality problem. What’s more, we proved that the technique of optimal step size for solving the variational inequality is still efficient for the split equality problem. Meanwhile, we also confirmed that these provided methods also apply to solving the split feasibility problem.

Generating Bézier curves for medical image reconstruction

The boundaries of medical images can be represented by a sequence of 2D curves from image slices taken from uncalibrated cameras. The 2D curves are used to reconstruct a 3D tissues volume in order to understand of anatomical functionality and morphological analysis. For that, new curves are needed to represent a sequence of 2D points sampled along the boundary.In this paper, we construct and analyse special types of Bézier curves for improving medical image reconstruction. The special types of Bézier curves are derived from the moving Frenet frames accompanying them. We describe and analysis of these curves in terms of their curvatures, torsions and control points. In addition, we introduce some important theories regarding the geometric properties of these curves. The thoughtful work represents the first attempt to deal with constructing such curves and might deliver fundamentals which are an important input in the algorithm for creating 3D surfaces. We provided an example with drawing as application, through which we demonstrate confirmation of our theoretical results.

Feature Extraction and Reconstruction of Medical Images using Two-Dimensional Principal Component Analysis

Two-Dimensional Principal Component Analysis (2DPCA) is a classical technique used to reduce the cost of computation than standard PCA. In 2DPCA, images are treated as vectors and appear image as a matrix which is further computed and results as Eigenvalues consisting of lower dimensionality as compared to PCA. 2DPCA depends on the image matrix which is a computationally most efficient method than PCA used to enhance feature extraction speed and high accuracy. 2DPCA is represented as an image matrix, its Co-Variance matrix is computed with the image matrix directly without converting into a 1D vector, and Eigenvectors are obtained for feature extraction. 2DPCA computes an accurate Co-Variance matrix and finds Eigenvectors most efficiently. K-Nearest Neighbor (KNN) algorithm is used for classification. 2DPCA is the best method to obtain reconstruction accuracy than PCA. The main advantage of 2DPCA is less time required for feature extraction and to provide the highest reconstruction accuracy. For testing and evaluating 2DPCA performance, we are conducted several experiments using different databases such as IRMA, WANG, etc., for different medical images and observed that reconstruction accuracy depends on increasing the number of principal components.

Efficient solution of linear inverse problems using an iterative linear neural network with a generalization training approach

Inverse problems are some of the most important mathematical problems in science and mathematics because their solution yields information about parameters that are not directly observable. Artificial neural networks have long been used as a mathematical modelling method and have been used successfully to solve inverse problems for application including denoising and medical image reconstruction. Many inverse problems result from integral processes that can be modelled using a linear formulation. These can be efficiently solved via simple networks which are easily trained with reasonable datasets. An innovative simple neural network architecture, the iterative linear neural network (ILNN), consisting of two non-hidden layer networks, one for the forward model and one for the inverse model, is proposed to solve linear inverse problems. Iteration between the two models refines network outcomes with greater accuracy than a network with only the inverse model. A training procedure accompanying the network is also introduced. The network needs to train only the inverse model with one-hot vectors as targets. The training inputs of the inverse model define the weights of the forward model. The number of targets is finite and equal to the length of the vector. With the defined targets, the training process ensures that the inverse model is at least a left inverse of the forward model. This leads to generalizable networks. The experimental results show that the ILNN produces good results even if its inverse model is not perfectly trained. The proposed network is applied to solve two linear inverse problems, deconvolution and the inverse Radon transform. The network successfully reconstructed original data following blurring and Radon transformation.

Reconstruction, processing and smoothing of surface geometry of a patient specific ascending aortic aneurysm

This work presents the creation of geometric surface models representing the heart chambers and the aorta for two patient specific cases. The first one concerns an acute ascending aortic aneurysm in the anterior Valsalva sinus and the second one a normal aorta for comparison reasons. The surface modelling was implemented in a medical image reconstruction software (Materialise Mimics), based on the computed tomography scans of the patients. For each case, the two atriums and ventricles, the ascending aorta, aortic arch and descending aorta, the pulmonary artery and the superior vena cava were also reconstructed and modelled. Special attention is given to the morphology of the vessels and the heart chambers that are in contact with the ascending aortic aneurysm, due to the acute aneurysmal extension of the Valsalva sinus. A comparison of the non-aneurysmal and the aneurysmal models is necessary to infer whether this disease affects the geometries and/or other vital organs that come in contact with the aortic aneurysm. Special care is given to the surfaces’ smoothing process to preserve the morphology of the structures, avoiding any deviations from the actual geometries of the corresponding vessels and heart chambers.

Scalable signal reconstruction for a broad range of applications

Signal reconstruction problem (SRP) is an important optimization problem where the objective is to identify a solution to an underdetermined system of linear equations that is closest to a given prior. It has a substantial number of applications in diverse areas, such as network traffic engineering, medical image reconstruction, acoustics, astronomy, and many more. Unfortunately, most of the common approaches for solving SRP do not scale to large problem sizes. We propose a novel and scalable algorithm for solving this critical problem. Specifically, we make four major contributions. First, we propose a dual formulation of the problem and develop the DIRECT algorithm that is significantly more efficient than the state of the art. Second, we show how adapting database techniques developed for scalable similarity joins provides a substantial speedup over DIRECT. Third, we describe several practical techniques that allow our algorithm to scale---on a single machine---to settings that are orders of magnitude larger than previously studied. Finally, we use the database techniques of materialization and reuse to extend our result to dynamic settings where the input to the SRP changes. Extensive experiments on real-world and synthetic data confirm the efficiency, effectiveness, and scalability of our proposal.

DUG-RECON: A Framework for Direct Image Reconstruction Using Convolutional Generative Networks

This article explores convolutional generative networks as an alternative to iterative reconstruction algorithms in medical image reconstruction. The task of medical image reconstruction involves mapping of projection domain data collected from the detector to the image domain. This mapping is done typically through iterative reconstruction algorithms which are time consuming and computationally expensive. Trained deep learning networks provide faster outputs as proven in various tasks across computer vision. In this work, we propose a direct reconstruction framework exclusively with deep learning architectures. The proposed framework consists of three segments, namely, denoising, reconstruction, and super resolution (SR). The denoising and the SR segments act as processing steps. The reconstruction segment consists of a novel double U-Net generator (DUG) which learns the sinogram-to-image transformation. This entire network was trained on positron emission tomography (PET) and computed tomography (CT) images. The reconstruction framework approximates 2-D mapping from the projection domain to the image domain. The architecture proposed in this proof-of-concept work is a novel approach to direct image reconstruction; further improvement is required to implement it in a clinical setting.

Reconstruction of High Resolution Medical Image Using General Regression Neural Network (GRNN)

Low image resolution has deficiencies in the diagnostic process, this will affect the quality of the image in describing an object in certain tissues or organs, especially in the process of examining patients by doctors or physicians based on the results of imaging medical devices such as CT-scans, MRIs and X-rays. Therefore, this study had developed a General Regression Neural Network (GRNN) type artificial neural network system to reconstruct a medical image so that the image has a significant resolution for the analysis process. The GRNN input layer uses grayscale intensity values with variations in the image position coordinates to produce an optimal resolution. There are four layers in this method, the first is input layer, the second is hidden layer, the third is summation, and the last layer is output. We examined the two parameters with different interval values of 0.2 and of 0.5. The result shows that the interval value of 0.2 is the optimal value to produce an output image that is identical to the input image. This is also supported by the results of the intensity curve of the RGB pattern matched between target and output.

Virtuel 3D simulering vs. traditionel anatomiundervisning

Fremskridt inden for IT-baseret medicinsk billedrekonstruktion har gjort billedbehandling i såvel 2D som 3D til en vigtig del af billeddiagnostikken. Dette stiller særlige krav til faggrupper som f.eks. radiografer. I dette projekt har vi kvantitativt målt radiografstuderendes læring af 3D-anatomi ved virtuel 3D simulering på Sectra Table og sammenlignet denne med læringsudbyttet ved traditionel anatomiundervisning. To kohorter radiografstuderende udgjorde studiepopulationen. 1. kohorte gennemgik et undervisningsforløb baseret på simulering i 3D. 2. kohorte gennemgik et forløb af tilsvarende længde med traditionel holdundervisning. De studerendes viden om 3D-anatomi blev testet før og efter intervention. Ændringer i testresultater sammenlignedes mellem de to kohorter. Studiet har ikke påvist statistisk signifikante forskelle i radiografstuderendes læringsudbytte af 3D-anatomi mellem de to undervisningsmetoder. Undersøgelsens resultater tyder dog på, at Sectra Table kan være et udmærket supplement til traditionel holdundervisning, når hensigten er, at radiografstuderende skal lære 3D-anatomi.

Rethinking medical image reconstruction via shape prior, going deeper and faster: Deep joint indirect registration and reconstruction.

Indirect image registration is a promising technique to improve image reconstruction quality by providing a shape prior for the reconstruction task. In this paper, we propose a novel hybrid method that seeks to reconstruct high quality images from few measurements whilst requiring low computational cost. With this purpose, our framework intertwines indirect registration and reconstruction tasks is a single functional. It is based on two major novelties. Firstly, we introduce a model based on deep nets to solve the indirect registration problem, in which the inversion and registration mappings are recurrently connected through a fixed-point interaction based sparse optimisation. Secondly, we introduce specific inversion blocks, that use the explicit physical forward operator, to map the acquired measurements to the image reconstruction. We also introduce registration blocks based deep nets to predict the registration parameters and warp transformation accurately and efficiently. We demonstrate, through extensive numerical and visual experiments, that our framework outperforms significantly classic reconstruction schemes and other bi-task method; this in terms of both image quality and computational time. Finally, we show generalisation capabilities of our approach by demonstrating their performance on fast Magnetic Resonance Imaging (MRI), sparse view computed tomography (CT) and low dose CT with measurements much below the Nyquist limit.

Laser Medical Image Reconstruction and Computer Aided Design of Polymethyl Methacrylate for Pediatric Removable Space Maintainer Applications

Background: Although there are many advantages of removable space maintainers (RSM), current methods of making RSM have some limitations. CAD/CAM technology may be used to solve these problems. Methods: Digital model of dentition defects were obtained using scanning technique coupled with laser medical image reconstruction. The digital RSM were designed using 3 shape software. They were manufactured using two methods: Polymethyl methacrylate (PMMA), and conventional methods. The spaces between the tissue surfaces of the RSM and the models were replaced using silicone, the maximum, mean distances and standard deviation were measured. A three-dimensional variation analysis was used to measure these spaces. The variance test compares the difference in distances among various materials. Results: The PMMA RSMs fit the models well. The maximum and mean distances of the PMMA groups were significantly smaller than those of conventional group (p &lt; 0.01). Furthermore, there were no significant different for the standard deviations among PMMA group and the conventional group. Conclusion: Digitally designed and integrated RSMs have good suitability and were better than that of conventional method. PMMA manufactured RSM using CAD/CAM technique may be applied to clinical.

Medical image segmentation and reconstruction of prostate tumor based on 3D AlexNet

BackgroundProstate cancer is a disease with a high incidence of tumors in men. Due to the long incubation time and insidious condition, early diagnosis is difficult; especially imaging diagnosis is more difficult. In actual clinical practice, the method of manual segmentation by medical experts is mainly used, which is time-consuming and labor-intensive and relies heavily on the experience and ability of medical experts. The rapid, accurate and repeatable segmentation of the prostate area is still a challenging problem. It is important to explore the automated segmentation of prostate images based on the 3D AlexNet network. MethodTaking the medical image of prostate cancer as the entry point, the three-dimensional data is introduced into the deep learning convolutional neural network. This paper proposes a 3D AlexNet method for the automatic segmentation of prostate cancer magnetic resonance images, and the general network ResNet 50, Inception -V4 compares network performance. ResultsBased on the training samples of magnetic resonance images of 500 prostate cancer patients, a set of 3D AlexNet with simple structure and excellent performance was established through adaptive improvement on the basis of classic AlexNet. The accuracy rate was as high as 0.921, the specificity was 0.896, and the sensitivity It is 0.902 and the area under the receiver operating characteristic curve (AUC) is 0.964. The Mean Absolute Distance (MAD) between the segmentation result and the medical expert's gold standard is 0.356 mm, and the Hausdorff distance (HD) is 1.024 mm, the Dice similarity coefficient is 0.9768. ConclusionThe improved 3D AlexNet can automatically complete the structured segmentation of prostate magnetic resonance images. Compared with traditional segmentation methods and depth segmentation methods, the performance of the 3D AlexNet network is superior in terms of training time and parameter amount, or network performance evaluation. Compared with the algorithm, it proves the effectiveness of this method.

Corrigendum to “Reconstruction of Medical Images Using Artificial Bee Colony Algorithm”

Corrigendum to “Reconstruction of Medical Images Using Artificial Bee Colony Algorithm”

A Novel Ray-Casting Algorithm Using Dynamic Adaptive Sampling

Ray-casting algorithm is an important volume rendering algorithm, which is widely used in medical image processing. Aiming to address the shortcomings of the current ray-casting algorithms in 3D reconstruction of medical images, such as slow rendering speed and low sampling efficiency, an improved algorithm based on dynamic adaptive sampling is proposed. By using the central difference gradient method, the corresponding sampling interval is obtained dynamically according to the different sampling points. Meanwhile, a new rendering operator is proposed based on the color value and opacity changes before and after the ray enters the volume element, and the resistance luminosity. Compared with the state of other algorithms, experimental results show that the method proposed in this paper has a faster rendering speed while ensuring the quality of the generated image.

Deep Medical Image Reconstruction with Autoencoders using Deep Boltzmann Machine Training

INTRODUCTION: Deep learning-based Image compression achieves a promising result in recent years as compared with the traditional transform coding methodology. Autoencoder, an unsupervised learning algorithm with the input value as same as that of the output value, is considered in this research work for effective medical image reconstruction.OBJECTIVES: Medical data needs to be reconstructed without distorting the details present over it. A deep neural network that accepts the data and processes it to the other several layers and reconstructs that data is achieved by autoencoder.METHODS: Deep Autoencoder is implemented in this methodology as it has been considered for high dimensionality reduction. Layer by layer pretraining is achieved using an approximate inference algorithm called Deep Boltzmann Machine.RESULTS: The proposed method proves to be efficient when compared with the performance of the other autoencoders such as Deep Autoencoder with multiple Backpropagation (DA-MBP), Deep Autoencoder with RBM (DA-RBM) and Deep Convolutional Autoencoder with RBM (DCA-RBM).CONCLUSION: Performance metrics are measured in terms of Mean Square Error, Structural similarity Index and PSNR.

Immersive Surgical Anatomy of the Craniocervical Junction.

With the advent and increased usage of posterior, lateral, and anterior surgical approaches to the craniocervical junction (CCJ), it is essential to have a sound understanding of the osseous, ligamentous, and neurovascular layers of this region as well as their three-dimensional (3D) orientations and functional kinematics. Advances in 3D technology can be leveraged to develop a more nuanced and comprehensive understanding of the CCJ, classically depicted via dissections and sketches. As such, this study aims to illustrate - with the use of 3D technologies - the major anatomical landmarks of the CCJ in an innovative and informative way. Photogrammetry, structured light scanning, and 3D reconstruction of medical images were used to generate these high-resolution volumetric models. A clear knowledge of the critical anatomical structures and morphometrics of the CCJ is crucial for the diagnosis, classification, and treatment of pathologies in this transitional region.