Robust Spectral Clustering Using Statistical Sub-Graph Affinity Model

Spectral clustering methods have been shown to be effective for image segmentation. Unfortunately, the presence of image noise as well as textural characteristics can have a significant negative effect on the segmentation performance. To accommodate for image noise and textural characteristics, this study introduces the concept of sub-graph affinity, where each node in the primary graph is modeled as a sub-graph characterizing the neighborhood surrounding the node. The statistical sub-graph affinity matrix is then constructed based on the statistical relationships between sub-graphs of connected nodes in the primary graph, thus counteracting the uncertainty associated with the image noise and textural characteristics by utilizing more information than traditional spectral clustering methods. Experiments using both synthetic and natural images under various levels of noise contamination demonstrate that the proposed approach can achieve improved segmentation performance when compared to existing spectral clustering methods.     

A Review on Convolutional Neural Networks for Brain Tumor Segmentation: Methods,Datasets, Libraries,and Future Directions

ObjectivesAccurate and reliable segmentation of brain tumors from MRI images helps in planning an enhanced treatment and increases the life expectancy of patients. However, the manual segmentation of brain tumors is subjective and more prone to errors. Nonetheless, the recent advances in convolutional neural network (CNN)-based methods have exhibited outstanding potential in robust segmentation of brain tumors. This article comprehensively investigates recent advances in CNN-based methods for automatic segmentation of brain tumors from MRI images. It examines popular deep learning (DL) libraries/tools for an expeditious and effortless implementation of CNN models. Furthermore, a critical assessment of current DL architectures is delineated along with the scope of improvement.MethodsIn this work, more than 50 scientific papers from 2014-2020 are selected using Google Scholar and PubMed. Also, the leading journals related to our work along with proceedings from major conferences such as MICCAI, MIUA and ECCV are retrieved. This research investigated various annual challenges too related to this work including Multimodal Brain Tumor Segmentation Challenge (MICCAI BRATS) and Ischemic Stroke Lesion Segmentation Challenge (ISLES).ResultAfter a systematic literature search pertinent to the theme, we found that principally there exist three variations of CNN architecture for brain tumor segmentation: single-path and multi-path, fully convolutional, and cascaded CNNs. The respective performances of most automated methods based on CNN are appraised on the BraTS dataset, provided as a part of the MICCAI Multimodal Brain Tumor Segmentation challenge held annually since 2012.ConclusionNotwithstanding the remarkable potential of CNN-based methods, reliable and robust segmentation of brain tumors continues to be an intractable challenge. This is due to the intricate anatomy of the brain, variability in its appearance, and imperfection in image acquisition. Moreover, owing to the small size of MRI datasets, CNN-based methods cannot operate with their full capacity, as demonstrated with large scale datasets, such as ImageNet.     

Rough-Fuzzy Clustering and Unsupervised Feature Selection for Wavelet Based MR Image Segmentation

Image segmentation is an indispensable process in the visualization of human tissues, particularly during clinical analysis of brain magnetic resonance (MR) images. For many human experts, manual segmentation is a difficult and time consuming task, which makes an automated brain MR image segmentation method desirable. In this regard, this paper presents a new segmentation method for brain MR images, integrating judiciously the merits of rough-fuzzy computing and multiresolution image analysis technique. The proposed method assumes that the major brain tissues, namely, gray matter, white matter, and cerebrospinal fluid from the MR images are considered to have different textural properties. The dyadic wavelet analysis is used to extract the scale-space feature vector for each pixel, while the rough-fuzzy clustering is used to address the uncertainty problem of brain MR image segmentation. An unsupervised feature selection method is introduced, based on maximum relevance-maximum significance criterion, to select relevant and significant textural features for segmentation problem, while the mathematical morphology based skull stripping preprocessing step is proposed to remove the non-cerebral tissues like skull. The performance of the proposed method, along with a comparison with related approaches, is demonstrated on a set of synthetic and real brain MR images using standard validity indices.     

A Split-and-Merge-Based Uterine Fibroid Ultrasound Image Segmentation Method in HIFU Therapy

High-intensity focused ultrasound (HIFU) therapy has been used to treat uterine fibroids widely and successfully. Uterine fibroid segmentation plays an important role in positioning the target region for HIFU therapy. Presently, it is completed by physicians manually, reducing the efficiency of therapy. Thus, computer-aided segmentation of uterine fibroids benefits the improvement of therapy efficiency. Recently, most computer-aided ultrasound segmentation methods have been based on the framework of contour evolution, such as snakes and level sets. These methods can achieve good performance, although they need an initial contour that influences segmentation results. It is difficult to obtain the initial contour automatically; thus, the initial contour is always obtained manually in many segmentation methods. A split-and-merge-based uterine fibroid segmentation method, which needs no initial contour to ensure less manual intervention, is proposed in this paper. The method first splits the image into many small homogeneous regions called superpixels. A new feature representation method based on texture histogram is employed to characterize each superpixel. Next, the superpixels are merged according to their similarities, which are measured by integrating their Quadratic-Chi texture histogram distances with their space adjacency. Multi-way Ncut is used as the merging criterion, and an adaptive scheme is incorporated to decrease manual intervention further. The method is implemented using Matlab on a personal computer (PC) platform with Intel Pentium Dual-Core CPU E5700. The method is validated on forty-two ultrasound images acquired from HIFU therapy. The average running time is 9.54 s. Statistical results showed that SI reaches a value as high as 87.58%, and normHD is 5.18% on average. It has been demonstrated that the proposed method is appropriate for segmentation of uterine fibroids in HIFU pre-treatment imaging and planning.     

A novel multiphoton microscopy images segmentation method based on superpixel and watershed

Multiphoton microscopy (MPM) imaging technique based on two‐photon excited fluorescence (TPEF) and second harmonic generation (SHG) shows fantastic performance for biological imaging. The automatic segmentation of cellular architectural properties for biomedical diagnosis based on MPM images is still a challenging issue. A novel multiphoton microscopy images segmentation method based on superpixels and watershed (MSW) is presented here to provide good segmentation results for MPM images. The proposed method uses SLIC superpixels instead of pixels to analyze MPM images for the first time. The superpixels segmentation based on a new distance metric combined with spatial, CIE Lab color space and phase congruency features, divides the images into patches which keep the details of the cell boundaries. Then the superpixels are used to reconstruct new images by defining an average value of superpixels as image pixels intensity level. Finally, the marker‐controlled watershed is utilized to segment the cell boundaries from the reconstructed images. Experimental results show that cellular boundaries can be extracted from MPM images by MSW with higher accuracy and robustness.

     

Improved SLIC imagine segmentation algorithm based on K-means

Dividing the image into superpixels contributes to further processing of the image. Simple linear iterative clustering (SLIC) algorithm achieves good segmentation result by clustering color and distance characteristics of pixels. However, finite superpixels easily cause under-segmentation. Therefore, the work corrects segmentation result of SLIC by k-means clustering method calculating similarity based on weighted Euclidean distance. After that, the under-segmentation superpixel blocks are conducted with k-means clustering based on binary classification. Result shows that the corrected SLIC segmentation has better visual effect and index.     

SegNet and Salp Water Optimization-driven Deep Belief Network for Segmentation and Classification of Brain Tumor

Classification of brain tumor in Magnetic Resonance Imaging (MRI) images is highly popular in treatment planning, early diagnosis, and outcome evaluation. It is very difficult for classifying and diagnosing tumors from several images. Thus, an automatic prediction strategy is essential in classifying brain tumors as malignant, core, edema, or benign. In this research, a novel approach using Salp Water Optimization-based Deep Belief network (SWO-based DBN) is introduced to classify brain tumor. At the initial stage, the input image is pre-processed to eradicate the artifacts present in input image. Following pre-processing, the segmentation is executed by SegNet, where the SegNet is trained using the proposed SWO. Moreover, the Convolutional Neural Network (CNN) features are employed to mine the features for future processing. At last, the introduced SWO-based DBN technique efficiently categorizes the brain tumor with respect to the extracted features. Thereafter, the produced output of the introduced SegNet + SWO-based DBN is made use of in brain tumor segmentation and classification. The developed technique produced better results with highest values of accuracy at 0.933, specificity at 0.880, and sensitivity at 0.938 using BRATS, 2018 datasets and accuracy at 0.921, specificity at 0.853, and sensitivity at 0.928 for BRATS, 2020 dataset.     

Quantitative Tumor Segmentation for Evaluation of Extent of Glioblastoma Resection to Facilitate Multisite Clinical Trials

Standard-of-care therapy for glioblastomas, the most common and aggressive primary adult brain neoplasm, is maximal safe resection, followed by radiation and chemotherapy. Because maximizing resection may be beneficial for these patients, improving tumor extent of resection (EOR) with methods such as intraoperative 5-aminolevulinic acid fluorescence-guided surgery (FGS) is currently under evaluation. However, it is difficult to reproducibly judge EOR in these studies due to the lack of reliable tumor segmentation methods, especially for postoperative magnetic resonance imaging (MRI) scans. Therefore, a reliable, easily distributable segmentation method is needed to permit valid comparison, especially across multiple sites. We report a segmentation method that combines versatile region-of-interest blob generation with automated clustering methods. We applied this to glioblastoma cases undergoing FGS and matched controls to illustrate the method's reliability and accuracy. Agreement and interrater variability between segmentations were assessed using the concordance correlation coefficient, and spatial accuracy was determined using the Dice similarity index and mean Euclidean distance. Fuzzy C-means clustering with three classes was the best performing method, generating volumes with high agreement with manual contouring and high interrater agreement preoperatively and postoperatively. The proposed segmentation method allows tumor volume measurements of contrast-enhanced T₁-weighted images in the unbiased, reproducible fashion necessary for quantifying EOR in multicenter trials.     

A Combinational Clustering Based Method for cDNA Microarray Image Segmentation

Microarray technology plays an important role in drawing useful biological conclusions by analyzing thousands of gene expressions simultaneously. Especially, image analysis is a key step in microarray analysis and its accuracy strongly depends on segmentation. The pioneering works of clustering based segmentation have shown that k-means clustering algorithm and moving k-means clustering algorithm are two commonly used methods in microarray image processing. However, they usually face unsatisfactory results because the real microarray image contains noise, artifacts and spots that vary in size, shape and contrast. To improve the segmentation accuracy, in this article we present a combination clustering based segmentation approach that may be more reliable and able to segment spots automatically. First, this new method starts with a very simple but effective contrast enhancement operation to improve the image quality. Then, an automatic gridding based on the maximum between-class variance is applied to separate the spots into independent areas. Next, among each spot region, the moving k-means clustering is first conducted to separate the spot from background and then the k-means clustering algorithms are combined for those spots failing to obtain the entire boundary. Finally, a refinement step is used to replace the false segmentation and the inseparable ones of missing spots. In addition, quantitative comparisons between the improved method and the other four segmentation algorithms--edge detection, thresholding, k-means clustering and moving k-means clustering--are carried out on cDNA microarray images from six different data sets. Experiments on six different data sets, 1) Stanford Microarray Database (SMD), 2) Gene Expression Omnibus (GEO), 3) Baylor College of Medicine (BCM), 4) Swiss Institute of Bioinformatics (SIB), 5) Joe DeRisi’s individual tiff files (DeRisi), and 6) University of California, San Francisco (UCSF), indicate that the improved approach is more robust and sensitive to weak spots. More importantly, it can obtain higher segmentation accuracy in the presence of noise, artifacts and weakly expressed spots compared with the other four methods.     

Machine Learning of Hierarchical Clustering to Segment 2D and 3D Images

We aim to improve segmentation through the use of machine learning tools during region agglomeration. We propose an active learning approach for performing hierarchical agglomerative segmentation from superpixels. Our method combines multiple features at all scales of the agglomerative process, works for data with an arbitrary number of dimensions, and scales to very large datasets. We advocate the use of variation of information to measure segmentation accuracy, particularly in 3D electron microscopy (EM) images of neural tissue, and using this metric demonstrate an improvement over competing algorithms in EM and natural images.     

A Context-Aware Delayed Agglomeration Framework for Electron Microscopy Segmentation

Electron Microscopy (EM) image (or volume) segmentation has become significantly important in recent years as an instrument for connectomics. This paper proposes a novel agglomerative framework for EM segmentation. In particular, given an over-segmented image or volume, we propose a novel framework for accurately clustering regions of the same neuron. Unlike existing agglomerative methods, the proposed context-aware algorithm divides superpixels (over-segmented regions) of different biological entities into different subsets and agglomerates them separately. In addition, this paper describes a “delayed” scheme for agglomerative clustering that postpones some of the merge decisions, pertaining to newly formed bodies, in order to generate a more confident boundary prediction. We report significant improvements attained by the proposed approach in segmentation accuracy over existing standard methods on 2D and 3D datasets.     

A diabetic retinopathy detection method using an improved pillar K-means algorithm

The paper presents a new approach for medical image segmentation. Exudates are a visible sign of diabetic retinopathy that is the major reason of vision loss in patients with diabetes. If the exudates extend into the macular area, blindness may occur. Automated detection of exudates will assist ophthalmologists in early diagnosis. This segmentation process includes a new mechanism for clustering the elements of high-resolution images in order to improve precision and reduce computation time. The system applies K-means clustering to the image segmentation after getting optimized by Pillar algorithm; pillars are constructed in such a way that they can withstand the pressure. Improved pillar algorithm can optimize the K-means clustering for image segmentation in aspects of precision and computation time. This evaluates the proposed approach for image segmentation by comparing with Kmeans and Fuzzy C-means in a medical image. Using this method, identification of dark spot in the retina becomes easier and the proposed algorithm is applied on diabetic retinal images of all stages to identify hard and soft exudates, where the existing pillar K-means is more appropriate for brain MRI images. This proposed system help the doctors to identify the problem in the early stage and can suggest a better drug for preventing further retinal damage.     

A Hybrid CNN-SVM Threshold Segmentation Approach for Tumor Detection and Classification of MRI Brain Images

ObjectiveIn this research paper, the brain MRI images are going to classify by considering the excellence of CNN on a public dataset to classify Benign and Malignant tumors.Materials and MethodsDeep learning (DL) methods due to good performance in the last few years have become more popular for Image classification. Convolution Neural Network (CNN), with several methods, can extract features without using handcrafted models, and eventually, show better accuracy of classification. The proposed hybrid model combined CNN and support vector machine (SVM) in terms of classification and with threshold-based segmentation in terms of detection.ResultThe findings of previous studies are based on different models with their accuracy as Rough Extreme Learning Machine (RELM)-94.233%, Deep CNN (DCNN)-95%, Deep Neural Network (DNN) and Discrete Wavelet Autoencoder (DWA)-96%, k-nearest neighbors (kNN)-96.6%, CNN-97.5%. The overall accuracy of the hybrid CNN-SVM is obtained as 98.4959%.ConclusionIn today's world, brain cancer is one of the most dangerous diseases with the highest death rate, detection and classification of brain tumors due to abnormal growth of cells, shapes, orientation, and the location is a challengeable task in medical imaging. Magnetic resonance imaging (MRI) is a typical method of medical imaging for brain tumor analysis. Conventional machine learning (ML) techniques categorize brain cancer based on some handicraft property with the radiologist specialist choice. That can lead to failure in the execution and also decrease the effectiveness of an Algorithm. With a brief look came to know that the proposed hybrid model provides more effective and improvement techniques for classification.     

Effects of non-local diffusion on structural MRI preprocessing and default network mapping: statistical comparisons with isotropic/anisotropic diffusion

Neuroimaging community usually employs spatial smoothing to denoise magnetic resonance imaging (MRI) data, e.g., Gaussian smoothing kernels. Such an isotropic diffusion (ISD) based smoothing is widely adopted for denoising purpose due to its easy implementation and efficient computation. Beyond these advantages, Gaussian smoothing kernels tend to blur the edges, curvature and texture of images. Researchers have proposed anisotropic diffusion (ASD) and non-local diffusion (NLD) kernels. We recently demonstrated the effect of these new filtering paradigms on preprocessing real degraded MRI images from three individual subjects. Here, to further systematically investigate the effects at a group level, we collected both structural and functional MRI data from 23 participants. We first evaluated the three smoothing strategies' impact on brain extraction, segmentation and registration. Finally, we investigated how they affect subsequent mapping of default network based on resting-state functional MRI (R-fMRI) data. Our findings suggest that NLD-based spatial smoothing maybe more effective and reliable at improving the quality of both MRI data preprocessing and default network mapping. We thus recommend NLD may become a promising method of smoothing structural MRI images of R-fMRI pipeline.     

Semi-automatic Method for Low-Grade Gliomas Segmentation in Magnetic Resonance Imaging

Background: Analyzing MR scans of low-grade glioma, with highly accurate segmentation will have an enormous potential in neurosurgery for diagnosis and therapy planning. Low-grade gliomas are mainly distinguished by their infiltrating character and irregular contours, which make the analysis, and therefore the segmentation task, more difficult. Moreover, MRI images show some constraints such as intensity variation and the presence of noise.Methods: To tackle these issues, a novel segmentation method built from the local properties of image is presented in this paper. Phase-based edge detection is estimated locally by the monogenic signal using quadrature filters. This way of detecting edges is, from a theoretical point of view, intensity invariant and responds well to the MR images. To strengthen the tumor detection process, a region-based term is designated locally in order to achieve a local maximum likelihood segmentation of the region of interest. A Gaussian probability distribution is considered to model local images intensities.Results: The proposed model is evaluated using a set of real subjects and synthetic images derived from the Brain Tumor Segmentation challenge –BraTS 2015. In addition, the obtained results are compared to the manual segmentation performed by two experts. Quantitative evaluations are performed using the proposed approach with regard to four related existing methods.Conclusion: The comparison of the proposed method, shows more accurate results than the four existing methods.     

Colorization and automated segmentation of human T2 MR brain images for characterization of soft tissues

Characterization of tissues like brain by using magnetic resonance (MR) images and colorization of the gray scale image has been reported in the literature, along with the advantages and drawbacks. Here, we present two independent methods; (i) a novel colorization method to underscore the variability in brain MR images, indicative of the underlying physical density of bio tissue, (ii) a segmentation method (both hard and soft segmentation) to characterize gray brain MR images. The segmented images are then transformed into color using the above-mentioned colorization method, yielding promising results for manual tracing. Our color transformation incorporates the voxel classification by matching the luminance of voxels of the source MR image and provided color image by measuring the distance between them. The segmentation method is based on single-phase clustering for 2D and 3D image segmentation with a new auto centroid selection method, which divides the image into three distinct regions (gray matter (GM), white matter (WM), and cerebrospinal fluid (CSF) using prior anatomical knowledge). Results have been successfully validated on human T2-weighted (T2) brain MR images. The proposed method can be potentially applied to gray-scale images from other imaging modalities, in bringing out additional diagnostic tissue information contained in the colorized image processing approach as described.     

A pipeline to quantify spinal cord atrophy with deep learning: Application to differentiation of MS and NMOSD patients

PurposeQuantitative measurement of various anatomical regions of the brain and spinal cord (SC) in MRI images are used as unique biomarkers to consider progress and effects of demyelinating diseases of the central nervous system. This paper presents a fully-automated image processing pipeline which quantifies the SC volume of MRI images.MethodsIn the proposed pipeline, after conducting some pre-processing tasks, a deep convolutional network is utilized to segment the spinal cord cross-sectional area (SCCSA) of each slice. After full segmentation, certain extra slices interpolate between each two adjacent slices using the shape-based interpolation method. Then, a 3D model of the SC is reconstructed, and, by counting the voxels of it, the SC volume is calculated. The performance of the proposed method for the SCCSA segmentation is evaluated on 140 MRI images. Subsequently, to demonstrate the application of the proposed pipeline, we study the differentiations of SC atrophy between 38 Multiple Sclerosis (MS) and 25 Neuromyelitis Optica Spectrum Disorder (NMOSD) patients.ResultsThe experimental results of the SCCSA segmentation indicate that the proposed method, adapted by Mask R-CNN, presented the most satisfactory result with the average Dice coefficient of 0.96. For this method, statistical metrics including sensitivity, specificity, accuracy, and precision are 97.51%, 99.98%, 99.92%, and 98.04% respectively. Moreover, the t-test result (p-value = 0.00089) verified a significant difference between the SC atrophy of MS and NMOSD patients.ConclusionThe pipeline efficiently quantifies the SC volume of MRI images and can be utilized as an affordable computer-aided tool for diagnostic purposes.     

A Probabilistic Atlas of Diffuse WHO Grade II Glioma Locations in the Brain

Diffuse WHO grade II gliomas are diffusively infiltrative brain tumors characterized by an unavoidable anaplastic transformation. Their management is strongly dependent on their location in the brain due to interactions with functional regions and potential differences in molecular biology. In this paper, we present the construction of a probabilistic atlas mapping the preferential locations of diffuse WHO grade II gliomas in the brain. This is carried out through a sparse graph whose nodes correspond to clusters of tumors clustered together based on their spatial proximity. The interest of such an atlas is illustrated via two applications. The first one correlates tumor location with the patient’s age via a statistical analysis, highlighting the interest of the atlas for studying the origins and behavior of the tumors. The second exploits the fact that the tumors have preferential locations for automatic segmentation. Through a coupled decomposed Markov Random Field model, the atlas guides the segmentation process, and characterizes which preferential location the tumor belongs to and consequently which behavior it could be associated to. Leave-one-out cross validation experiments on a large database highlight the robustness of the graph, and yield promising segmentation results.     

Spatial segmentation of imaging mass spectrometry data with edge-preserving image denoising and clustering

In recent years, matrix-assisted laser desorption/ionization (MALDI)-imaging mass spectrometry has become a mature technology, allowing for reproducible high-resolution measurements to localize proteins and smaller molecules. However, despite this impressive technological advance, only a few papers have been published concerned with computational methods for MALDI-imaging data. We address this issue proposing a new procedure for spatial segmentation of MALDI-imaging data sets. This procedure clusters all spectra into different groups based on their similarity. This partition is represented by a segmentation map, which helps to understand the spatial structure of the sample. The core of our segmentation procedure is the edge-preserving denoising of images corresponding to specific masses that reduces pixel-to-pixel variability and improves the segmentation map significantly. Moreover, before applying denoising, we reduce the data set selecting peaks appearing in at least 1% of spectra. High dimensional discriminant clustering completes the procedure. We analyzed two data sets using the proposed pipeline. First, for a rat brain coronal section the calculated segmentation maps highlight the anatomical and functional structure of the brain. Second, a section of a neuroendocrine tumor invading the small intestine was interpreted where the tumor area was discriminated and functionally similar regions were indicated.     

Accurate Image Analysis of the Retina Using Hessian Matrix and Binarisation of Thresholded Entropy with Application of Texture Mapping

In this paper, we demonstrate a comprehensive method for segmenting the retinal vasculature in camera images of the fundus. This is of interest in the area of diagnostics for eye diseases that affect the blood vessels in the eye. In a departure from other state-of-the-art methods, vessels are first pre-grouped together with graph partitioning, using a spectral clustering technique based on morphological features. Local curvature is estimated over the whole image using eigenvalues of Hessian matrix in order to enhance the vessels, which appear as ridges in images of the retina. The result is combined with a binarized image, obtained using a threshold that maximizes entropy, to extract the retinal vessels from the background. Speckle type noise is reduced by applying a connectivity constraint on the extracted curvature based enhanced image. This constraint is varied over the image according to each region''s predominant blood vessel size. The resultant image exhibits the central light reflex of retinal arteries and veins, which prevents the segmentation of whole vessels. To address this, the earlier entropy-based binarization technique is repeated on the original image, but crucially, with a different threshold to incorporate the central reflex vessels. The final segmentation is achieved by combining the segmented vessels with and without central light reflex. We carry out our approach on DRIVE and REVIEW, two publicly available collections of retinal images for research purposes. The obtained results are compared with state-of-the-art methods in the literature using metrics such as sensitivity (true positive rate), selectivity (false positive rate) and accuracy rates for the DRIVE images and measured vessel widths for the REVIEW images. Our approach out-performs the methods in the literature.