Medical Image Processing Applications Research Articles

Development of Power-Delay Product Optimized ASIC-Based Computational Unit for Medical Image Compression

The proliferation of battery-operated end-user electronic devices due to technological advancements, especially in medical image processing applications, demands low power consumption, high-speed operation, and efficient coding. The design of these devices is centered on the Application-Specific Integrated Circuits (ASIC), General Purpose Processors (GPP), and Field Programmable Gate Array (FPGA) frameworks. The need for low-power functional blocks arises from the growing demand for high-performance computational units that are part of high-speed processors operating at high clock frequencies. The operational speed of the processor is determined by the computational unit, which is the workhorse of high-speed processors. A novel approach to integrating Very Large-Scale Integration (VLSI) ASIC design and the concepts of low-power VLSI compatible with medical image compression was embraced in this research. The focus of this study was the design, development, and implementation of a Power Delay Product (PDP) optimized computational unit targeted for medical image compression using ASIC design flow. This stimulates the research community’s quest to develop an ideal architecture, emphasizing on minimizing power consumption and enhancing device performance for medical image processing applications. The study uses area, delay, power, PDP, and Peak Signal-to-Noise Ratio (PSNR) as performance metrics. The research work takes inspiration from this and aims to enhance the efficiency of the computational unit through minor design modifications that significantly impact performance. This research proposes to explore the trade-off of high-performance adder and multiplier designs to design an ASIC-based computational unit using low-power techniques to enhance the efficiency in power and delay. The computational unit utilized for the digital image compression process was synthesized and implemented using gpdk 45 nm standard libraries with the Genus tool of Cadence. A reduced PDP of 46.87% was observed when the image compression was performed on a medical image, along with an improved PSNR of 5.89% for the reconstructed image.

Application of machine learning methods and medical image processing in solving the problem of detecting stenoses of the middle cerebral artery according to computed tomographic angiography data

Application of machine learning methods and medical image processing in solving the problem of detecting stenoses of the middle cerebral artery according to computed tomographic angiography data

Exploring the Application of the Artificial-Intelligence-Integrated Platform 3D Slicer in Medical Imaging Education.

Artificial Intelligence (AI) has revolutionized medical imaging procedures, specifically with regard to image segmentation, reconstruction, interpretation, and research. 3D Slicer, an open-source medical image analysis platform, has become a valuable tool in medical imaging education due to its integration of various AI applications. Through its open-source architecture, students can gain practical experience with diverse medical images and the latest AI technology, reinforcing their understanding of anatomy and imaging technology while fostering independent learning and clinical reasoning skills. The implementation of this platform improves instruction quality and nurtures skilled professionals who can meet the demands of clinical practice, research institutions, and technology innovation enterprises. AI algorithms' application in medical image processing have facilitated their translation from the lab to practical clinical applications and education.

Transformers and their application to medical image processing: A review

Transformers perform well in natural language processing tasks and have made many breakthroughs in computer vision. In medical image processing, transformers are successfully used in image segmentation, classification, reconstruction, and diagnosis. In this paper, we mainly expound on the transformer principle and its application in medical imaging. Specifically, we first introduce the basic principles and model structure of transformers. Then, we summarize the improvement mechanism of the transformer's network including combining the transformer with the Unet network, creating a transformer lightweight variant network, strengthening the cross-fast link mechanism, and building a large model with the transformer as the skeleton. Second, extensive discussion is given to medical image segmentation, reconstruction, classification, and other applications. Finally, the main challenges transformers face in the medical image processing field and future development prospects. Furthermore, we systematically summarize the latest research progress of transformers and their application in medical image processing, which has significant reference value for transformer research in the medical field.

Low-dose COVID-19 CT Image Denoising Using CNN and its Method Noise Thresholding.

The novel coronavirus outbreak has ushered in different new areas of research in medical instrumentation and technology. Medical diagnostics and imaging are one of the ways in which the area and level of infection can be detected. COVID-19 attacks people with less immunity, so infants, kids, and pregnant women are more vulnerable to the infection. So, they need to undergo CT scanning to find the infection level. But the high radiation diagnostic is also fatal for them, so the intensity of radiation needs to be reduced significantly, which may generate the noise in the CT images. This paper introduces a new denoising technique for low-dose Covid-19 CT images using a convolution neural network (CNN) and noise-based thresholding method. The major concern of the methodology for reducing the risk associated with radiation while diagnosing. The results are evaluated visually and using standard performance metrics. From comparative analysis, it was observed that proposed works give better outcomes. The proposed low-dose COVID-19 CT image denoising model is therefore concluded to have a better potential to be effective in various pragmatic medical image processing applications in noise suppression and clinical edge preservation.

Image Super-Resolution using Generative Adversarial Networks with EfficientNetV2

The image super-resolution is utilized for the image transformation from low resolution to higher resolution to obtain more detailed information to identify the targets. The super-resolution has potential applications in various domains, such as medical image processing, crime investigation, remote sensing, and other image-processing application domains. The goal of the super-resolution is to obtain the image with minimal mean square error with improved perceptual quality. Therefore, this study introduces the perceptual loss minimization technique through efficient learning criteria. The proposed image reconstruction technique uses the image super-resolution generative adversarial network (ISRGAN), in which the learning of the discriminator in the ISRGAN is performed using the EfficientNet-v2 to obtain a better image quality. The proposed ISRGAN with the EfficientNet-v2 achieved a minimal loss of 0.02, 0.1, and 0.015 at the generator, discriminator, and self-supervised learning, respectively, with a batch size of 32. The minimal mean square error and mean absolute error are 0.001025 and 0.00225, and the maximal peak signal-to-noise ratio and structural similarity index measure obtained are 45.56985 and 0.9997, respectively.

Automatic Polyp Detection by Combining Conditional Generative Adversarial Network and Modified You-Only-Look-Once

Recent years have seen deep learning algorithms such as Convolutional Neural Networks (CNNs) and their variants achieving competitive performance in the application of medical image processing. Yet their shortcomings are: (i) the need of a large collection of annotated data, and (ii) the involvement of careful design of CNN layers. In this paper, we presented a novel polyp detection architecture based on two ideas: (i) to use Conditional Generative Adversarial Network (CGAN) to expand the training datasets, specifically the Generator takes advantage of the Efficient Spatial Pyramid (ESP) and the Discriminator is based on the PatchGAN; (ii) to modify the architecture of YOLOv4 using dilated convolution and skip connections. Experiments were performed using three publicly available datasets, i.e., CVC-ClinicDB, CVC-ColonDB, and ETIS-Larib Polyp DB. Experimental results showed that our generative network outperformed U-Net and can synthesize more realistic polyp images. The modified YOLOv4 significantly improved the performance of polyp detection using the expanded dataset, with an accuracy of 92.37% and a detection rate of 17.2 frames per second.

A dense R‐CNN multi‐target instance segmentation model and its application in medical image processing

A dense R‐CNN multi‐target instance segmentation model and its application in medical image processing

Computer-Aided Diagnosis for Determining Sagittal Spinal Curvatures Using Deep Learning and Radiography.

Analyzing spinal curvatures manually is time-consuming and tedious for clinicians, and intra-observer and inter-observer variability can affect manual measurements. In this study, we developed and evaluated the performance of an automated deep learning-based computer-aided diagnosis (CAD) tool for measuring the sagittal alignment of the spine from X-ray images. The CAD system proposed here performs two functions: deep learning-based lateral spine segmentation and automatic analysis of thoracic kyphosis and lumbar lordosis angles. We utilized 322 datasets with data augmentation for learning and fivefold cross-validation. The segmentation model was based on U-Net, which has multiple applications in medical image processing. Here, we utilized parameter equations and trigonometric functions to design spinal angle measurement algorithms. The kyphosis (T4-T12) and lordosis angle (L1-S1, L1-L5) were automatically measured to help diagnose kyphosis and lordosis. The segmentation model had precision, sensitivity, and dice similarity coefficient values of 90.53 ± 4.61%, 89.53 ± 1.8%, and 90.22 ± 0.62%, respectively. The performance of the CAD algorithm was also verified with the Pearson correlation, Bland-Altman, and intra-class correlation coefficient (ICC) analysis. The proposed angle measurement algorithm exhibited high similarity and reliability during verification. Therefore, CAD can help clinicians in reaching a diagnosis by analyzing the sagittal spinal curvatures while reducing observer-based variability and the required time or effort.

Finite Radon Transform as Subcarriers Mapping Technique for OFDM System

Radon Transform (RT) becomes a mathematical basis for solving computerized tomography, and finds application in medical imaging and image processing. Finite groups represent digital images; the Finite Radon Transform (FRAT) is then applied. To overcome the periodization effect of a finite transform, Do and Vetterli [1] introduce a novel ordering of the FRAT coefficients. This paper introduced a detailed mathematical analysis of the optimal ordering of FRAT coefficients that introduces new normal vectors that are clearly identified with examples producing the FRAT transform. Lately FRAT is used with OFDM systems as an M-Ary mapping technique, thus, the signal constellation of some of the different orders of FRAT is presented as well as generating the OFDM symbols from the output of FRAT. This procedure leads to extra investigation on the performance of FRAT-based OFDM systems in comparison with the QAM mapping techniques.

Enhancement of White Blood Cells Images using Shock Filtering Equation for Classification Problem

Medical image processing has developed rapidly in the last decade. The autodetection and classification of white blood cells (WBC) is one of the medical image processing applications. The analysis of WBC images has engaged researchers from medical also technology fields. Since WBC detection plays an essential role in the medical field, this paper presents a system for distinguishing and classifying WBC types: eosinophils, neutrophils, lymphocytes, and monocytes, using K-Nearest Neighbor (K-NN) and Logistic Regression (LR). This study aims to find the best accuracy of pre-processing images using original grayscale, shock filtering, and thresholding grayscale. The highest average accuracy in classifying WBC images in the conducting research is 43.54% using the LR algorithm from 2103 images. It is obtained from the combination of thresholding grayscale image and shock filtering equation to enhance the quality of an image. Overall, using two algorithms, KNN and LR, the classification accuracy can increase up to 12%.

Automatic Diagnosis of Alzheimer’s Disease Using Deep Learning Techniques

Neuroimaging data analysis has attracted a great deal of attention with respect to the accurate diagnosis of Alzheimer’s disease (AD). Magnetic Resonance Imaging (MRI) scanners have thus been commonly used to study AD-related brain structural variations, providing images that demonstrate both morphometric and anatomical changes in the human brain. Deep learning algorithms have already been effectively exploited in other medical image processing applications to identify features and recognise patterns for many diseases that affect the brain and other organs; this paper extends on this to describe a novel computer aided software pipeline for the classification and early diagnosis of AD. The proposed method uses two types of three-dimensional Convolutional Neural Networks (3D CNN) to facilitate brain MRI data analysis and automatic feature extraction and classification, so that pre-processing and post-processing are utilised to normalise the MRI data and facilitate pattern recognition. The experimental results show that the proposed approach achieves 97.5%, 82.5%, and 83.75% accuracy in terms of binary classification AD vs. cognitively normal (CN), CN vs. mild cognitive impairment (MCI) and MCI vs. AD, respectively, as well as 85% accuracy for multi class-classification, based on publicly available data sets from the Alzheimer’s disease Neuroimaging Initiative (ADNI).

Comparative Analysis of Deep Convolutional Neural Network Models for Humerus Bone Fracture Detection

Deep learning plays a key role in medical image processing. One of the applications of deep learning models in this domain is bone fracture detection from X-ray images. Convolutional neural network and its variants are used in wide range of medical image processing applications. MURA Dataset is commonly used in various studies that detect bone fractures and this work also uses that dataset, in specific the Humerus bone radiograph images. The humerus dataset in the MURA dataset contains both images with fracture and without fracture. The image with fracture includes images with metals which are removed in this work. Experimental analysis was made with two variants of convolutional neural network, DenseNet169 Model and the VGG Model. In case of the DenseNet169 model, a model with the pre trained weights of ImageNet and one without it is experimented. Results obtained with these variants of CNN are comparedand it shows that DenseNet169 model that uses pre-trained weights of ImageNet model performs better than the other two models.

Pixel Distribution Learning for Vessel Segmentation under Multiple Scales.

In this work we try to address if there is a better way to classify two distributions, rather than using histograms; and answer if we can make a deep learning network learn and classify distributions automatically. These improvements can have wide ranging applications in computer vision and medical image processing. More specifically, we propose a new vessel segmentation method based on pixel distribution learning under multiple scales. In particular, a spatial distribution descriptor named Random Permutation of Spatial Pixels (RPoSP) is derived from vessel images and used as the input to a convolutional neural network for distribution learning. Based on our preliminary experiments we currently believe that a wide network, rather than a deep one, is better for distribution learning. There is only one convolutional layer, one rectified linear layer and one fully connected layer followed by a softmax loss in our network. Furthermore, in order to improve the accuracy of the proposed approach, the RPoSP features are captured at multiple scales and combined together to form the input of the network. Evaluations using standard benchmark datasets demonstrate that the proposed approach achieves promising results compared to the state-of-the-art.

Comparing Popular CNN Models for an Imbalanced Dataset of Dermoscopic Images

In this study, the performance of popular convolution architectures against an imbalanced dataset is analyzed in detail with a multi-classing medical image processing application. Our selection for dermoscopic images is a large-scale and imbalanced dataset consisting of 10,015 colored lesion images belonging to 7 different skin diseases, was used as a benchmark. Images without pathological testing are labeled by specialist dermatologists who are members of International Skin Imaging Association. The f1-score was preferred as the measurement metric during the training phase of the convolution networks, which were trained with imbalanced dataset, and the area under the receiver operating characteristic curve and the confusion matrix of each model were calculated at the test phase. In the validation phase of convolution networks, k-fold cross validation technique was used. In addition, the filters obtained from the ImageNet dataset have been imported with the Transfer-Learning option. Fine-tuning was applied to the deepest convolution layers in order for these pre-trained models to develop themselves specific to our application. In order to prevent the overfit problem, the feature extraction outputs of the models were drop-out at a rate of 50% after flattening, and L2-regularization (weigh decay) was applied during the update phase of the weights. Although it is not the main purpose of the study, in order to partially improve the performance of convolution architectures, synthetic lesion images created with data-augmentation for the minor classes in the imbalanced dataset were included in the training process in a way that does not cause information leakage.

A Two-Phase Approach using Mask R-CNN and 3D U-Net for High-Accuracy Automatic Segmentation of Pancreas in CT Imaging

Background and objectiveThe size, shape, and position of the pancreas are affected by the patient characteristics such as age, sex, adiposity. Owing to more complex anatomical structures (size, shape, and position) of the pancreas, specialists have some difficulties in the analysis of pancreatic diseases (diabetes, pancreatic cancer, pancreatitis). Therefore, the treatment of the disease requires enormous time and depends on the experience of specialists. In order to decrease the rate of pancreatic disease deaths and to assist the specialist in the analysis of pancreatic diseases, automatic pancreas segmentation techniques have been actively developed in the research article for many years. MethodsAlthough the rapid growth of deep learning and proving satisfactory performance in many medical image processing and computer vision applications, the maximum Dice Similarity Coefficients (DSC) value of these techniques related to automatic pancreas segmentation is only around 85% due to complex structure of the pancreas and other factors. Contrary to previous techniques which are required significantly higher computational power and memory, this paper suggests a novel two-phase approach for high-accuracy automatic pancreas segmentation in computed tomography (CT) imaging. The proposed approach consists of two phases; (1) Pancreas Localization, where the rough pancreas position is detected on the 2D CT slice by adopting Mask R-CNN model, (2) Pancreas Segmentation, where the segmented pancreas region is produced by refining the candidate pancreas region with 3D U-Net on the 2D sub-CT slices generated in the first phase. Both qualitative and quantitative assessments of the proposed approach are performed on the NIH data set. ResultsIn order to evaluate the achievement of the recommended approach, a total of 16 different automatic pancreas segmentation techniques reported in the literature are compared by utilizing performance assessment procedures which are Dice Similarity Coefficient (DSC), Jaccard Index (JI), Precision (PRE), Recall (REC), Pixel Accuracy (ACC), Specificity (SPE), Receiver Operating Characteristics (ROC) and Area under ROC curve (AUC). The average values of DSC, JI, REC and ACC are computed as 86.15%, 75.93%, 86.27%, 86.27% and 99.95% respectively, which are the best values among well-established studies for automatic pancreas segmentation. ConclusionIt is demonstrated with qualitative and quantitative results that our suggested two-phase approach creates more favorable results than other existing approaches.

Jacobsthal Matrices and their Properties

Objectives: Matrices with Jacobsthal numbers are used in the medical image processing applications. The Cholesky factorization of the matrix with the Jacobsthal number is anlayzed. We also investigate the upper and lower bounds of the eigenvalues of the symmetric Jacobsthal and Jacobsthal-Lucas matrices. Methods: In this paper, we define a factor matrix and use the factorization techniques to get Cholesky decomposition of the Jacobsthal, Jacobsthal-Lucas matrix and inverses of these matrices. The bounds for eigenvalues are obtained using majorization techniques. Findings: The Cholesky factorization has been obtained using the factor matrix technique for any matrix of order n with entries from the Jacobsthal and Jacobsthal Lucas sequences. Novelty: Factorization of Lucas and symmetric Lucas matrix has already been obtained using the factorization technique. In this paper we give the factorization of the matrices with entries from the Jacobsthal and Jacobsthal Lucas sequences. Mathematics Subject Classification (2020). 15A23, 11B39, 15A18,15A42 Keywords: Jacobsthal matrix; Jacobsthal-Lucas matrix; symmetric; eigenvalues

TMD-Unet: Triple-Unet with Multi-Scale Input Features and Dense Skip Connection for Medical Image Segmentation.

Deep learning is one of the most effective approaches to medical image processing applications. Network models are being studied more and more for medical image segmentation challenges. The encoder–decoder structure is achieving great success, in particular the Unet architecture, which is used as a baseline architecture for the medical image segmentation networks. Traditional Unet and Unet-based networks still have a limitation that is not able to fully exploit the output features of the convolutional units in the node. In this study, we proposed a new network model named TMD-Unet, which had three main enhancements in comparison with Unet: (1) modifying the interconnection of the network node, (2) using dilated convolution instead of the standard convolution, and (3) integrating the multi-scale input features on the input side of the model and applying a dense skip connection instead of a regular skip connection. Our experiments were performed on seven datasets, including many different medical image modalities such as colonoscopy, electron microscopy (EM), dermoscopy, computed tomography (CT), and magnetic resonance imaging (MRI). The segmentation applications implemented in the paper include EM, nuclei, polyp, skin lesion, left atrium, spleen, and liver segmentation. The dice score of our proposed models achieved 96.43% for liver segmentation, 95.51% for spleen segmentation, 92.65% for polyp segmentation, 94.11% for EM segmentation, 92.49% for nuclei segmentation, 91.81% for left atrium segmentation, and 87.27% for skin lesion segmentation. The experimental results showed that the proposed model was superior to the popular models for all seven applications, which demonstrates the high generality of the proposed model.

Medical image segmentation using customized U-Net with adaptive activation functions

Since medical imaging is a fundamental step in clinical diagnosis and treatment, medical image processing is an attractive field for researchers. Among the different applications of medical image processing, this paper focuses on the segmentation task using a customized deep convolutional neural network (CNN). The proposed network is developed based on the idea of improving a deep network performance and speeding up its learning process while using less parameters. Using the famous U-Net architecture which has proven its effectiveness in the segmentation field, the customization is done here by applying adaptive activation functions. In the proposed network, the U-Net complexity is reduced about 200 times by alleviating some parameters to accelerate the learning process. However, the important modification is the use of adaptive activation functions where each convolution layer learns its own data-adaptive activation function as a linear combination of 16 well-known basic functions. This modification successfully compensates the accuracy drop caused by parameter alleviation and also makes the model capable to be tuned with small amount of training data. Conducting several experiments on five famous retinal image datasets, namely DRIVE, STARE, CHASE, HRF, and ARIA, the proposed customized U-Net achieved 96%, 97%, 96%, 97%, and 95% accuracy in segmenting blood vessels, respectively. The proposed network also showed 98% accuracy on ISIC skin lesion dataset for segmenting the lesion area. The obtained results obviously confirm the high performance of the proposed customized network compared to the previous successful researches in handling the medical segmentation task. They also light the hope that many famous deep networks can benefit from these types of customization to become efficient compact models with the ability to handle lack of sufficient data.

Hybrid Manta Ray Foraging Optimization for Novel Brain Tumor Detection

In medical image processing, segmentation and extraction of tumor portion from brain MRI is a complex task. It consumes more time and human effort to differentiate the normal and abnormal tissue. Clinical experts need more time to provide accurate results, recent technology developments in image processing reduces the human effort and provides more accurate results which reduces time and death rates by identifying the issues in early stage itself. Machine learning based algorithms occupies a major role in bio medical image processing applications. The performance of machine learning models is in satisfactory levels, but it could be improved by introducing optimization in feature selection stage itself. The research work provides a hybrid manta ray foraging optimization for feature selection from brain tumor MRI images. Convolution neural network is used to test the optimized features and detects the early stage brain tumors. The experimental model is compared with existing artificial neural network, particle swarm optimization algorithm and acquires a better detection and classification accuracy.