Medical CT Images Research Articles

Brain CT and MRI medical image fusion using convolutional neural networks and a dual-channel spiking cortical model.

The aim of medical image fusion is to improve the clinical diagnosis accuracy, so the fused image is generated by preserving salient features and details of the source images. This paper designs a novel fusion scheme for CT and MRI medical images based on convolutional neural networks (CNNs) and a dual-channel spiking cortical model (DCSCM). Firstly, non-subsampled shearlet transform (NSST) is utilized to decompose the source image into a low-frequency coefficient and a series of high-frequency coefficients. Secondly, the low-frequency coefficient is fused by the CNN framework, where weight map is generated by a series of feature maps and an adaptive selection rule, and then the high-frequency coefficients are fused by DCSCM, where the modified average gradient of the high-frequency coefficients is adopted as the input stimulus of DCSCM. Finally, the fused image is reconstructed by inverse NSST. Experimental results indicate that the proposed scheme performs well in both subjective visual performance and objective evaluation and has superiorities in detail retention and visual effect over other current typical ones. Graphical abstract A schematic diagram of the CT and MRI medical image fusion framework using convolutional neural network and a dual-channel spiking cortical model.

Image segmentation algorithm based on geometric flow Bandelets transformation particle replanting

In order to solve the problem that the pixel inhomogeneity of the medical CT image has a great influence on the image local segmentation algorithm, a new image segmentation algorithm based on the geometric flow Bandelets transformation particle replanting is proposed. Firstly, in view of the inhomogeneity of the local image, the embedded transition interface is reconstructed by calculating the Lagrangian labeled particles before calculating the particle replanting formula, thus improving the quality conservation characteristics of the particle replanting algorithm. Secondly, in view of the uncertainty of the traditional particle method in dealing with the singularity of the interface and the complex geometric related problems, the velocity vector and unit normal vector are added to promote the convergence of the velocity field of singular point and the topological change point. Finally, the simulation test on the synthetic data test set and the real CT image shows that the proposed algorithm is superior to the contrast algorithm in the convergence accuracy and operation speed of the edge segmentation.

Ripplet domain fusion approach for CT and MR medical image information

Multimodal medical image fusion (MIF) plays an important role as an assistant for medical professionals by providing a better visualization of diagnostic information using different imaging modalities. The process of image fusion helps the radiologists in the precise diagnosis of several critical diseases and its treatment. In this paper, the proposed framework presents a fusion approach for multimodal medical images that utilize both the features extracted by the discrete ripplet transform (DRT) and pulse coupled neural network. The DRT having different features and a competent depiction of the image coefficients provides several directional high-frequency subband coefficients. The DRT decomposition can preserve more detailed information present in the reference images and further enhance the visualization of the fused images. Firstly, the DRT is applied to decompose the reference images into several low and high-frequency subimage coefficients that are fused by computing the novel sum modified Laplacian and novel modified spatial frequency motivated pulse coupled neural model. This model is used to preserve the redundant information also. Finally, fused images are reconstructed by applying the inverse DRT. The performance of the proposed fusion approach is validated by extensive simulation on the different CT-MR image datasets. Experimental results demonstrate that the proposed method provides the better fused images in terms of visual quality along with the quantitative measures as compared to several existing fusion approaches.

Convolutional neural network-based multimodal image fusion via similarity learning in the shearlet domain

Recently, deep learning has been shown effectiveness in multimodal image fusion. In this paper, we propose a fusion method for CT and MR medical images based on convolutional neural network (CNN) in the shearlet domain. We initialize the Siamese fully convolutional neural network with a pre-trained architecture learned from natural data; then, we train it with medical images in a transfer learning fashion. Training dataset is made of positive and negative patch pair of shearlet coefficients. Examples are fed in two-stream deep CNN to extract features maps; then, a similarity metric learning based on cross-correlation is performed aiming to learn mapping between features. The minimization of the logistic loss objective function is applied with stochastic gradient descent. Consequently, the fusion process flow starts by decomposing source CT and MR images by the non-subsampled shearlet transform into several subimages. High-frequency subbands are fused based on weighted normalized cross-correlation between feature maps given by the extraction part of the CNN, while low-frequency coefficients are combined using local energy. Training and test datasets include pairs of pre-registered CT and MRI taken from the Harvard Medical School database. Visual analysis and objective assessment proved that the proposed deep architecture provides state-of-the-art performance in terms of subjective and objective assessment. The potential of the proposed CNN for multi-focus image fusion is exhibited in the experiments.

Can we trust the calculation of texture indices of CT images? A phantom study.

Texture analysis is an emerging tool in the field of medical imaging analysis. However, many issues have been raised in terms of its use in assessing patient images and it is crucial to harmonize and standardize this new imaging measurement tool. This study was designed to evaluate the reliability of texture indices of CT images on a phantom including a reproducibility study, to assess the discriminatory capacity of indices potentially relevant in CT medical images and to determine their redundancy. For the reproducibility and discriminatory analysis, eight identical CT acquisitions were performed on a phantom including one homogeneous insert and two close heterogeneous inserts. Texture indices were selected for their high reproducibility and capability of discriminating different textures. For the redundancy analysis, 39 acquisitions of the same phantom were performed using varying acquisition parameters and a correlation matrix was used to explore the 2 × 2 relationships. LIFEx software was used to explore 34 different parameters including first order and texture indices. Only eight indices of 34 exhibited high reproducibility and discriminated textures from each other. Skewness and kurtosis from histogram were independent from the six other indices but were intercorrelated, the other six indices correlated in diverse degrees (entropy, dissimilarity, and contrast of the co-occurrence matrix, contrast of the Neighborhood Gray Level difference matrix, SZE, ZLNU of the Gray-Level Size Zone Matrix). Care should be taken when using texture analysis as a tool to characterize CT images because changes in quantitation may be primarily due to internal variability rather than from real physio-pathological effects. Some textural indices appear to be sufficiently reliable and capable to discriminate close textures on CT images.

Improving Medical CT Image Blind Restoration Algorithm Based on Dictionary Learning by Alternating Direction Method of Multipliers

In this paper, the medical CT image blind restoration is translated into two sub problems, namely, image estimation based on dictionary learning and point spread function estimation. A blind restoration algorithm optimized by the alternating direction method of multipliers for medical CT images was proposed. At present, the existing methods of blind image restoration based on dictionary learning have the problem of low efficiency and precision. This paper aims to improve the effectiveness and accuracy of the algorithm and to improve the robustness of the algorithm. The local CT images are selected as training samples, and the K-SVD algorithm is used to construct the dictionary by iterative optimization, which is beneficial to improve the efficiency of the algorithm. Then, the orthogonal matching pursuit algorithm is employed to implement the dictionary update. Dictionary learning is accomplished by sparse representation of medical CT images. The alternating direction method of multipliers (ADMM) is used to solve the objective function and realize the local image restoration, so as to eliminate the influence of point spread function. Secondly, the local restoration image is used to estimate the point spread function, and the convex quadratic optimization method is used to solve the point spread function sub problems. Finally, the optimal estimation of point spread function is obtained by iterative method, and the global sharp image is obtained by the alternating direction method of multipliers. Experimental results show that, compared with the traditional adaptive dictionary restoration algorithm, the new algorithm improves the objective image quality metrics, such as peak signal to noise ratio, structural similarity, and universal image quality index. The new algorithm optimizes the restoration effect, improves the robustness of noise immunity and improves the computing efficiency.

Biomechanical Analysis of Hip Joint Arthroplasties using CT-Image Based Finite Element Method

In the present study, medical CT images were utilized to develop 3D-computational models of hip joints and femoral bones. The finite element analysis was then performed to characterize the biomechanical problems associated with the bone models. The FEA results clearly exhibited dramatic changes of stress and SED profiles due to osteoarthritis and hip arthroplasty. Furthermore, a damage modelling was also introduced into the FEA to characterize the fracture problems of femoral bone with prostheses under a downstairs step and sideways fall conditions. Micro-damage formations are well predicted in the vicinity of the prostheses for bone cases. It is thus confirmed that CT-FEA can be very useful to understand the biomechanical problems in the field of orthopaedics.

The Architectural Design of Deblocking Filter for Image Enhancement in the Diagnosis of Cancer

Objective:This method is to reduce the risk of imprecise diagnosis associated with poor-quality CT images, this paper presents a new technique designed to enhance the quality of medical CT images. The main objective is to improve the appearance of CT images in order to obtain better visual interpretation and analysis, which is expected to ease the diagnosis process. The proposed technique involves applying a deblocking filter is to enhance the visual nature of a picture by diminishing the blocking artifacts. The appearance of a picture isn’t clear while an antique happens. The proposed deblocking filter calculation gives a strategy to expel the ancient rarities by smoothing the sharp edges of a picture.Methods:With a specific goal to lessen the quantity of information access, multifaceted nature and consequently to upgrade the proficiency, a six-staged pipelined structure for picture pixels are proposed. Besides, to enhance the subjective and target nature of a picture the deblocking filter performs identification of the antique at the coded square limits and weakens them by applying a chose filter.Result:The proposed algorithm is implemented in HDL using Xilinx FPGA. The input image is converted into decimal pixel values using Matlab and this value is used as the input in HDL. The proposed algorithm is compared with other blocking algorithms.Conclusion:To design an effective deblocking filter with low cost, low complexity and high intensity, pipeline based systems are used. In addition to that the number of memory accesses and timing efficiency also be reduced using this method. The deblocking filtering operations can also easily perform in parallel on multiple processors by using six-stage of pipelined, two-line deblocking filter. The parameter mean, variance, standard deviation, resolution, contrast and PSNR values are compared with the previous method. Hence it shows the implementation of deblocking filter using pipelining is more efficient than others.

Nonsubsampled shearlet domain fusion techniques for CT–MR neurological images using improved biological inspired neural model

The fusion of multimodality medical images performs a very crucial role in the clinical diagnosis, analysis and the treatment of especially in critical diseases. It is considered as an assisted approach for the radiologist by providing the composite images having significant diagnostic information acquired from the source images. The main purpose of this work is to develop an efficient framework for fusing the multimodal medical images. Three different fusion techniques are proposed in this paper that presents the CT and MR medical image fusion in nonsubsampled shearlet transform (NSST) domain using the adaptive spiking neural model. The NSST having different features and a competent depiction of the image coefficients provides several directional decomposition coefficients. Maximum selection approach and regional energy are utilized for low frequency coefficients fusion. Spatial frequency, novel modified spatial frequency and novel sum modified Laplacian motivated spiking model are used for every high frequency subimage component. Finally, fused images are reconstructed by applying inverse NSST. The performance of proposed fusion techniques is validated by extensive simulations performed on different CT-MR image datasets using proposed and other thirty seven existing fusion approaches in terms of both the subjective and objective manner. The results revealed that the proposed techniques provide better visualization of resultant images and higher quantitative measures compared to several existing fusion approaches.

Validation of Material Algorithms for Femur Remodelling Using Medical Image Data.

The aim of this study is the utilization of human medical CT images to quantitatively evaluate two sorts of “error-driven” material algorithms, that is, the isotropic and orthotropic algorithms, for bone remodelling. The bone remodelling simulations were implemented by a combination of the finite element (FE) method and the material algorithms, in which the bone material properties and element axes are determined by both loading amplitudes and daily cycles with different weight factor. The simulation results showed that both algorithms produced realistic distribution in bone amount, when compared with the standard from CT data. Moreover, the simulated L-T ratios (the ratio of longitude modulus to transverse modulus) by the orthotropic algorithm were close to the reported results. This study suggests a role for “error-driven” algorithm in bone material prediction in abnormal mechanical environment and holds promise for optimizing implant design as well as developing countermeasures against bone loss due to weightlessness. Furthermore, the quantified methods used in this study can enhance bone remodelling model by optimizing model parameters to gap the discrepancy between the simulation and real data.

A New Adaptive Gamma Correction Based Algorithm Using DWT-SVD for Non-Contrast CT Image Enhancement.

The performances of medical image processing techniques, in particular CT scans, are usually affected by poor contrast quality introduced by some medical imaging devices. This suggests the use of contrast enhancement methods as a solution to adjust the intensity distribution of the dark image. In this paper, an advanced adaptive and simple algorithm for dark medical image enhancement is proposed. This approach is principally based on adaptive gamma correction using discrete wavelet transform with singular-value decomposition (DWT-SVD). In a first step, the technique decomposes the input medical image into four frequency sub-bands by using DWT and then estimates the singular-value matrix of the low-low (LL) sub-band image. In a second step, an enhanced LL component is generated using an adequate correction factor and inverse singular value decomposition (SVD). In a third step, for an additional improvement of LL component, obtained LL sub-band image from SVD enhancement stage is classified into two main classes (low contrast and moderate contrast classes) based on their statistical information and therefore processed using an adaptive dynamic gamma correction function. In fact, an adaptive gamma correction factor is calculated for each image according to its class. Finally, the obtained LL sub-band image undergoes inverse DWT together with the unprocessed low-high (LH), high-low (HL), and high-high (HH) sub-bands for enhanced image generation. Different types of non-contrast CT medical images are considered for performance evaluation of the proposed contrast enhancement algorithm based on adaptive gamma correction using DWT-SVD (DWT-SVD-AGC). Results show that our proposed algorithm performs better than other state-of-the-art techniques.

A simple texture feature for retrieval of medical images

Texture characteristic is an important attribute of medical images, and has been applied in many medical image applications. This paper proposes a simple approach to employ the texture features of medical images for retrieval. The developed approach first conducts image filtering to medical images using different Gabor and Schmid filters, and then uniformly partitions the filtered images into non-overlapping patches. These operations provide extensive local texture information of medical images. The bag-of-words model is finally used to obtain feature representations of the images. Compared with several existing features, the proposed one is more discriminative and efficient. Experiments on two benchmark medical CT image databases have demonstrated the effectiveness of the proposed approach.

Double regularization medical CT image blind restoration reconstruction based on proximal alternating direction method of multipliers

To solve the problem of CT image degradation, a double regularization CT image blind restoration reconstruction method was proposed. The objective function including both a clear image and point spread function was established. To avoid the over-smoothing phenomenon and protect the detail, the objective function includes two constraint regularization terms. They are total variation (TV) and wavelet sparsity respectively. The objective function was solved by the alternating direction multiplier method (ADMM), and the optimal solution was obtained. Firstly, the CT image blind restoration reconstruction was decomposed into two sub-problems: reconstructed image estimation and point spread function estimation. Furthermore, each sub-problem can be solved by the proximal alternating direction method of multipliers. Finally, the CT image blind restoration reconstruction was realized. The experimental results show that the proposed algorithm takes into account the degradation effect of projection data, and the proposed algorithm is superior to other existing algorithms in the subjective visual effect. At the same time, in the aspect of objective evaluation, the proposed algorithm improves the objective image quality metrics such as peak signal-to-noise ratio (PSNR), structural similarity index metric (SSIM), and universal image quality index (UIQI).

An efficient and robust approach for biomedical image retrieval using Zernike moments

Success of any image retrieval system depends heavily on the feature extraction capability of its feature descriptor. In this paper, we present a biomedical image retrieval system which uses Zernike moments (ZMs) for extracting features from CT and MRI medical images. ZMs belong to the class of orthogonal rotation invariant moments (ORIMs) and possess very useful characteristics such as superior information representation capability with minimum redundancy, insensitivity to image noise etc. Existence of these properties as well as the ability of lower order ZMs to discriminate between different image shapes and textures motivated us to explore ZMs for biomedical retrieval application. To prove the effectiveness of our system, experiments have been carried out on both noise-free and noisy versions of two different medical databases i.e. Emphysema-CT database for CT image retrieval and OASIS-MRI database for MRI image retrieval. The proposed ZMs-based approach has been compared with the existing and recently published approaches based on local binary pattern (LBP), local ternary patterns (LTP), local diagonal extrema pattern (LDEP), etc., in terms of various evaluation measures like ARR, ARP, F _ score, and mAP. The results after being investigated have shown a significant improvement (10–14% and 15–17% in case of noise-free and noisy images, respectively) in comparison to the state-of-the-art techniques on the respective databases.

Migration mode of brine and supercritical CO2 during steady-state relative permeability measurements at very slow fluid flow velocity

Characteristics of two-phase flow for a brine–CO2 system in Berea sandstone were investigated to understand flow mechanisms on the basis of ‘ganglion dynamics’ and ‘connected-pathway flow’, both recently established through studies using synchrotron-beam micro-X-ray CT (μCT). Relative permeability curves (RPCs) in the brine–CO2 system were measured under steady-state conditions by calculating CO2 saturation from medical X-ray CT images. Fluid flow velocity was less than 10–5 m s−1, which represents the scenario of fluid movements near CO2 diffusion fronts in reservoirs. The flow direction was perpendicular to sedimentary layers during the injections in the experiment. The differential pressure across the core at each steady-flow state generally decreases with increasing CO2 content. This suggests a CO2 flow regime transition from one where ganglion dynamics dominated to one where connected-pathway-flow dominated. Randomly distributed pathways appear in each layer, resulting in a critical CO2 saturation (35–40 per cent) for achieving steady flow. The critical saturation value corresponds to the critical values reported in electrical resistivity in sandstones where the resistivity deviates from Archie's law. In contrast, for a flow direction parallel to sedimentary layers, steady-flow states are achieved when CO2 percolation clusters (channels) appear at lower CO2 saturation values (0–10 per cent). The flow perpendicular to bedding showed high CO2 saturation (ca. 40 per cent) at the brine-rich endpoint of RPC. Then, a rapid increase in RPC appeared at the CO2 saturation values 40–50 per cent. This suggests that the steady flow between inlet and outlet of sample ends is achieved through percolation pathways, and that the percolation clusters grow rapidly above critical saturation, as was indicated by the classical percolation theory. The relationship between sedimentary layers and flow direction seems to control the shapes of RPC, but the crucial factor is which type of CO2 pathway dominates: channeled pathway or randomly distributed pathway.

The diagnosis and treatment of systemic lupus erythematosus complicated with hydrocephalus

Objective: To investigate the pathogenesis and management of the hydrocephalus in patients with systemic lupus erythematosus (SLE). Methods: Eight cases of hospitalized lupus patients with hydrocephalus in Peking Union Medical College Hospital from Jan 1990 to Mar 2017 were analyzed retrospectively.We collected the patients' medical records including medical history, CT and/or MRI images and analysis of cerebrospinal fluid via lumbar puncture in order to investigate the pathogenesis of the hydrocephalus in patients with SLE and summarize treatment experience. Results: All the 8 patients with SLE complicated with hydrocephalus were treated with steroids or immunosuppressive agents, and 5 cases were diagnosed with the central nervous system infection. Three cases received ventriculoperitoneal shunt, 5 cases received treatment of medicine.They all were followed up for 2-12 months, of which 3 cases were markedly effective, 1 case effective and 4 cases dead. Conclusion: The pathogenesis of lupus combined with hydrocephalus may have a certain relationship with the central infection, should take effective anti-infection treatment.We recommend cerebrospinal fluid shunt surgery intervention.

A non invasive computer aided diagnosis system for early detection of lung carcinoma in ct medical images

A non invasive computer aided diagnosis system for early detection of lung carcinoma in ct medical images

Visual Three-Dimensional Reconstruction of Aortic Dissection Based on Medical CT Images

With the rapid development of CT technology, especially the higher resolution of CT machine and a sharp increase in the amount of slices, to extract and three-dimensionally display aortic dissection from the huge medical image data became a challenging task. In this paper, active shape model combined with spatial continuity was adopted to realize automatic reconstruction of aortic dissection. First, we marked aortic feature points from big data sample library and registered training samples to build a statistical model. Meanwhile, gray vectors were sampled by utilizing square matrix, which set the landmarks as the center. Posture parameters of the initial shape were automatically adjusted by the method of spatial continuity between CT sequences. The contrast experiment proved that the proposed algorithm could realize accurate aorta segmentation without selecting the interested region, and it had higher accuracy than GVF snake algorithm (93.29% versus 87.54% on aortic arch, 94.30% versus 89.25% on descending aorta). Aortic dissection membrane was extracted via Hessian matrix and Bayesian theory. Finally, the three-dimensional visualization of the aortic dissection was completed by volume rendering based on the ray casting method to assist the doctors in clinical diagnosis, which contributed to improving the success rate of the operations.

SU‐C‐207B‐01: A Novel Graphics Processing Units (GPU) Implementation of Discrete Wavelet Transformation

Purpose:To design and implement a GPU‐based discrete wavelet transformation (DWT) to be used in medical image reconstruction, image processing, and data compression. DWTs are widely used in medical physics, but the computation of DWTs is time consuming for large volumetric data. An efficient parallel implementation of DWTs is essential for many time sensitive applications, such as 4DCT.Methods:We choose Daubechies wavelet transformations as a benchmark, implementing both DWT and inverse DWT (IDWT). The reference CPU code is from “Numerical Recipes in C”. We implemented GPU‐based codes using C++ with CUDA 7.5. A GPU is specialized processors with a highly parallel structure originally designed for manipulating computer graphics. GPU computation is highly memory‐bounded. To optimize GPU memory access pattern and achieve best performance for transformation along Y‐direction, the data are transposed before and after the DWT (IDWT), which can sharply reduce computing efforts. Both CPU and GPU codes are unit‐tested and the result differences between two implementations are verified to be within rounding error. The hardware platform was a desktop with Intel Xeon E5‐1607 3.00 GHz CPU and NVIDIA Quadro K2000 GPU.Results:With the GPU implemented code, we process a medical CT image size of 512×512×256 in 0.5 seconds, which is 60 times faster than the CPU implementation. A naive non‐optimized GPU implementation by direct parallelization approaches only 10 times speedup for 2‐D and 3‐D DWT and IDWT. In comparison, the optimization of GPU memory access pattern can obtain an extra 6x speedup.Conclusion:We developed an efficient implementation of GPU‐based DWT/IDWT for medical image processing. To achieve the best performance of any medical imaging processing algorithm, developers should pay attention to memory access pattern optimization when implementing on GPU architecture.This project is supported by CPRIT grant under RP150485.

Cone beam computed tomography in sialography-report of two cases

Sialolithiasis is the most common obstructive conditions of the major salivary glands, among which submandibular salivary gland calculi are more prevalent. Various imaging modalities are routinely practiced these days for diagnosing sialolithiasis. Today, the fast acquisition of a 3D image using cone beam CT (CBCT) has enabled us to overcome the temporal limitations of medical CT and MRI image acquisition. An initial and precise diagnosis will aid in an effective treatment. We present a series of 2 case reports wherein CBCT sialography was performed to demonstrate the obstructive salivary gland diseases for better interpretation of the ductal system of the gland. CBCT promises to modernize the practice of oral and maxillofacial stomatologist by offering foremost advantages like high resolution, less scan time dose to and the patient.