Magnetic resonance imaging segmentation techniques using batch-type learning vector quantization algorithms

In this article, we propose batch-type learning vector quantization (LVQ) segmentation techniques for the magnetic resonance (MR) images. Magnetic resonance imaging (MRI) segmentation is an important technique to differentiate abnormal and normal tissues in MR image data. The proposed LVQ segmentation techniques are compared with the generalized Kohonen's competitive learning (GKCL) methods, which were proposed by Lin et al. [Magn Reson Imaging 21 (2003) 863-870]. Three MRI data sets of real cases are used in this article. The first case is from a 2-year-old girl who was diagnosed with retinoblastoma in her left eye. The second case is from a 55-year-old woman who developed complete left side oculomotor palsy immediately after a motor vehicle accident. The third case is from an 84-year-old man who was diagnosed with Alzheimer disease (AD). Our comparisons are based on sensitivity of algorithm parameters, the quality of MRI segmentation with the contrast-to-noise ratio and the accuracy of the region of interest tissue. Overall, the segmentation results from batch-type LVQ algorithms present good accuracy and quality of the segmentation images, and also flexibility of algorithm parameters in all the comparison consequences. The results support that the proposed batch-type LVQ algorithms are better than the previous GKCL algorithms. Specifically, the proposed fuzzy-soft LVQ algorithm works well in segmenting AD MRI data set to accurately measure the hippocampus volume in AD MR images.     

Iterative algorithm for spatial and intensity normalization of MEMRI images. Application to tract-tracing of rat olfactory pathways

Manganese (Mn)-enhanced magnetic resonance imaging (MEMRI) is an emerging technique for visualizing neuronal pathways and mapping brain activity modulation in animal models. Spatial and intensity normalizations of MEMRI images acquired from different subjects are crucial steps as they can influence the results of groupwise analysis. However, no commonly accepted procedure has yet emerged. Here, a normalization method is proposed that performs both spatial and intensity normalizations in a single iterative process without the arbitrary choice of a reference image. Spatial and intensity normalizations benefit from this iterative process. On one hand, spatial normalization increases the accuracy of region of interest (ROI) positioning for intensity normalization. On the other hand, improving the intensity normalization of the different MEMRI images leads to a better-averaged target on which the images are spatially registered. After automatic fast brain segmentation and optimization of the normalization process, this algorithm revealed the presence of Mn up to the posterior entorhinal cortex in a tract-tracing experiment on rat olfactory pathways. Quantitative comparison of registration algorithms showed that a rigid model with anisotropic scaling is the best deformation model for intersubject registration of three-dimensional MEMRI images. Furthermore, intensity normalization errors may occur if the ROI chosen for intensity normalization intersects regions where Mn concentration differs between experimental groups. Our study suggests that cross-comparing Mn-injected animals against a Mn-free group may provide a control to avoid bias introduced by intensity normalization quality. It is essential to optimize spatial and intensity normalization as the detectability of local between-group variations in Mn concentration is directly tied to normalization quality.     

An energy minimization method for MS lesion segmentation from T1-w and FLAIR images

In this paper, we extend the multiplicative intrinsic component optimization (MICO) algorithm to multichannel MR image segmentation, with focus on segmentation of multiple sclerosis (MS) lesions. The MICO algorithm was originally proposed by Li et al. in Ref. [1] for normal brain tissue segmentation and intensity inhomogeneity correction of a single channel MR image, which exhibits desirable advantages over other methods for MR image segmentation and intensity inhomogeneity correction in terms of segmentation accuracy and robustness. In this paper, we extend the MICO algorithm to multi-channel MR image segmentation and enable the segmentation of MS lesions. We assign different weights for different channels to control the impact of each channel. The weighted channels allow the enhancement of the impact of the FLAIR image on the segmentation of MS lesions by assigning a larger weight to the FLAIR image channel than the other channels. With the inherent mechanism of estimation of the bias field, our method is able to deal with the intensity inhomogeneity in the input multi-channel MR images. In the application of our method, we only use T1-w and FLAIR images as the input two channel MR images. Experimental results show promising result of our method.     

Generalized Kohonen's competitive learning algorithms for ophthalmological MR image segmentation

Kohonen's self-organizing map is a two-layer feedforward competitive learning network. It has been used as a competitive learning clustering algorithm. In this paper, we generalize Kohonen's competitive learning (KCL) algorithm with fuzzy and fuzzy-soft types called fuzzy KCL (FKCL) and fuzzy-soft KCL (FSKCL). These generalized KCL algorithms fuse the competitive learning with soft competition and fuzzy c-means (FCM) membership functions. We then apply these generalized KCLs to MRI and MRA ophthalmological segmentations. These KCL-based MRI segmentation techniques are useful in reducing medical image noise effects using a learning mechanism. They may be particularly helpful in clinical diagnosis. Two real cases with MR image data recommended by an ophthalmologist are examined. First case is a patient with Retinoblastoma in her left eye, an inborn malignant neoplasm of the retina frequently metastasis beyond the lacrimal cribrosa. The second case is a patient with complete left side oculomotor palsy immediately after a motor vehicle accident. Her brain MRI with MRA, skull routine, orbital CT, and cerebral angiography did not reveal brainstem lesions, skull fractures, or vascular anomalies. These generalized KCL algorithms were used in segmenting the ophthalmological MRIs. KCL, FKCL and FSKCL comparisons are made. Overall, the FSKCL algorithm is recommended for use in MR image segmentation as an aid to small lesion diagnosis.     

Workflow sensitivity of post-processing methods in renal DCE-MRI

Objective: Estimation of renal filtration using dynamic contrast-enhanced imaging (DCE-MRI) requires a series of analysis steps. The possible number of distinct post-processing chains is large and grows rapidly with increasing number of processing steps or options. In this study we introduce a framework for systematic evaluation of the post-processing chains. The framework is later used to highlight the workflow processing chain sensitivity towards accuracy in estimation of glomerular filtration rate (GFR).Methods: Twenty healthy volunteers underwent DCE-MRI examinations as well as iohexol clearance for reference GFR measurements. In total, 692 different combinations of post-processing steps were explored for analysis, including options for kidney segmentation, B1 inhomogeneity correction, placement of arterial input function, gadolinium concentration estimation as well as handling of motion-corrupted volumes and breathing motion. The evaluation of various processing chains is presented using a classification tree framework and random forest ensemble learning.Results: Among the processing steps subject to testing, methods for calculating the gadolinium concentration as well as B1 inhomogeneity correction had the largest impact on accuracy of GFR estimations. Different segmentation methods did not play an important role in the post-processing of the MR data except from one processing chain where the automated segmentation outperformed the manual segmentation.Conclusion: The proposed classification trees were efficiently used as a statistical tool for visualization and communication of results to distinguish between important and less influential processing steps in renal DCE-MRI. We also identified several crucial factors in the processing chain.     

Magnetic-resonance image segmentation based on improved variable weight multi-resolution Markov random field in undecimated complex wavelet domain

To solve the problem that the magnetic resonance (MR) image has weak boundaries, large amount of information, and low signal-to-noise ratio, we propose an image segmentation method based on the multi-resolution Markov random field (MRMRF) model. The algorithm uses undecimated dual-tree complex wavelet transformation to transform the image into multiple scales. The transformed low-frequency scale histogram is used to improve the initial clustering center of the K-means algorithm, and then other cluster centers are selected according to the maximum distance rule to obtain the coarse-scale segmentation. The results are then segmented by the improved MRMRF model. In order to solve the problem of fuzzy edge segmentation caused by the gray level inhomogeneity of MR image segmentation under the MRMRF model, it is proposed to introduce variable weight parameters in the segmentation process of each scale. Furthermore, the final segmentation results are optimized. We name this algorithm the variable-weight multi-resolution Markov random field (VWMRMRF). The simulation and clinical MR image segmentation verification show that the VWMRMRF algorithm has high segmentation accuracy and robustness, and can accurately and stably achieve low signal-to-noise ratio, weak boundary MR image segmentation.     

用于压缩感知磁共振成像的分割字典学习算法

字典学习算法可以根据数据本身的特点构建稀疏域中的基,从而使数据的表示更加稀疏.该文在传统的字典学习算法基础上提出了分割字典学习算法,由于部分磁共振图像组织结构简单、可以进行图像分割,因此可根据此特点来优化字典中基函数的构建,使磁共振图像的表达更为稀疏,从而获得更高的重建图像质量.该文利用模拟数据和真实数据进行了重建实验,结果表明与传统的字典学习算法相比,分割字典学习算法能进一步改善重建图像质量.     

Image-guided regularization level set evolution for MR image segmentation and bias field correction

Magnetic resonance (MR) image segmentation is a crucial step in surgical and treatment planning. In this paper, we propose a level-set-based segmentation method for MR images with intensity inhomogeneous problem. To tackle the initialization sensitivity problem, we propose a new image-guided regularization to restrict the level set function. The maximum a posteriori inference is adopted to unify segmentation and bias field correction within a single framework. Under this framework, both the contour prior and the bias field prior are fully used. As a result, the image intensity inhomogeneity can be well solved. Extensive experiments are provided to evaluate the proposed method, showing significant improvements in both segmentation and bias field correction accuracies as compared with other state-of-the-art approaches.     

Effect of nonrigid registration algorithms on deformation-based morphometry: a comparative study with control and Williams syndrome subjects

Deformation-based morphometry (DBM) is a widely used method for characterizing anatomical differences across groups. DBM is based on the analysis of the deformation fields generated by nonrigid registration algorithms, which warp the individual volumes to a DBM atlas. Although several studies have compared nonrigid registration algorithms for segmentation tasks, few studies have compared the effect of the registration algorithms on group differences that may be uncovered through DBM. In this study, we compared group atlas creation and DBM results obtained with five well-established nonrigid registration algorithms using 13 subjects with Williams syndrome and 13 normal control subjects. The five nonrigid registration algorithms include the following: (1) the adaptive bases algorithm, (2) the image registration toolkit, (3) The FSL nonlinear image registration tool, (4) the automatic registration tool, and (5) the normalization algorithm available in Statistical Parametric Mapping (SPM8). Results indicate that the choice of algorithm has little effect on the creation of group atlases. However, regions of differences between groups detected with DBM vary from algorithm to algorithm both qualitatively and quantitatively. Some regions are detected by several algorithms, but their extent varies. Others are detected only by a subset of the algorithms. Based on these results, we recommend using more than one algorithm when performing DBM studies.     

基于拉曼光谱技术的海水微塑料快速识别技术研究

近年来由于塑料的大量使用和排放,这些塑料经环境作用破碎变成微塑料大量汇聚到海洋中,导致海洋中聚集大量微塑料。微塑料形状较小,难以识别其来源与种类。激光拉曼探测技术具有快速、无损、且各物质指纹峰明显易被精确识别等优点,近年来被广泛应用。本文基于拉曼光谱探测技术,提出了一种结合小波处理、随机森林算法实现海水中微塑料快速识别的智能分类方法。针对六种典型的海水微塑料标准样品（丙烯腈(A)-丁二烯(B)-苯乙烯(S)的三元共聚物（ABS）、聚酰胺（PA）、聚对苯二甲酸乙二醇酯（PET）、聚丙烯（PP）、聚苯乙烯（PS）、聚氯乙烯（PVC））,采用激光拉曼探测技术进行光谱数据收集,对获取的拉曼光谱采用小波基为DB7、分解次数为3的小波,标准差归一化进行了拉曼光谱预处理。为了提高识别速度,同时还需要对光谱数据进行数据压缩预处理,分别进行了数据压缩点为64,128,256,512和1 024点的数据压缩比较,它们的决策树算法识别精度分别为91.51%,91.67%,92.35%,93.17%和93.21%,随机森林算法识别精度分别为93.12%,93.92%,94.83%,96.81%和96.81%,实验结果表明,微塑料的拉曼光谱压缩为512点时为效率和精度的最佳压缩点,可以为实际工程应用中微塑料拉曼数据压缩提供参考。分别采用决策树、随机森林两种算法进行微塑料拉曼光谱识别研究。研究结果表明,基于拉曼光谱数据,随机森林算法的识别微塑料交叉验证精度高于决策树算法。为进一步提高识别精度,进行了模型参数（折次k）优化研究,采用经过优化后的模型参数（k=20）,随机森林算法识别微塑料的交叉验证精度可以达到97.24%。可以为实际海水中微塑料的快速识别提供技术参考。     

An Interval Iteration Based Multilevel Thresholding Algorithm for Brain MR Image Segmentation

In this paper, we propose an interval iteration multilevel thresholding method (IIMT). This approach is based on the Otsu method but iteratively searches for sub-regions of the image to achieve segmentation, rather than processing the full image as a whole region. Then, a novel multilevel thresholding framework based on IIMT for brain MR image segmentation is proposed. In this framework, the original image is first decomposed using a hybrid L₁ − L₀ layer decomposition method to obtain the base layer. Second, we use IIMT to segment both the original image and its base layer. Finally, the two segmentation results are integrated by a fusion scheme to obtain a more refined and accurate segmentation result. Experimental results showed that our proposed algorithm is effective, and outperforms the standard Otsu-based and other optimization-based segmentation methods.     

Core level photoemission and STM characterization of Ta/Si(1 1 1)-7 × 7 interfaces

We present the results of scanning tunneling microscopy (STM) and photoemission spectroscopy (PES) of the Ta/Si(1 1 1)-7 × 7 system after deposition of Ta at substrate temperatures from 300 to 1250 K. The coverage of Ta varied from 0.05 up to 2.5 of a monolayer (ML). STM shows that at 300 K and coverage less than 1 ML, a disordered chemisorbed phase is formed. Deposition on a hot surface (above 500 K) produces round 3D clusters randomly distributed on the surface. Cluster height and their diameter are found to change drastically with annealing temperature and the Ta coverage. Analysis of photoemission data of the Si 2p core levels shows that at room temperature and at coverage ?1 ML core level binding energy shifts and intensity variations of Si surface related components are observed, which clearly indicate that the reaction starts already at 300 K. Shifts in the binding energy, changes of the peak shapes and intensity of the Ta 4f doublet at higher temperatures can be explained by the formation of stable silicide on the surface.     

Whole volume brain extraction for multi-centre,multi-disease FLAIR MRI datasets

Automatic segmentation of the brain from magnetic resonance images (MRI) is a fundamental step in many neuroimaging processing frameworks. There are mature technologies for this task for T1- and T2-weighted MRI; however, a widely-accepted brain extraction method for Fluid-Attenuated Inversion Recovery (FLAIR) MRI has yet to be established. FLAIR MRI are becoming increasingly important for the analysis of neurodegenerative diseases and tools developed for this sequence would have clinical value. To maximize translation opportunities and for large scale research studies, algorithms for brain extraction in FLAIR MRI should generalize to multi-centre (MC) data. To this end, this work proposes a fully automated, whole volume brain extraction methodology for MC FLAIR MRI datasets. The framework is built using a novel standardization framework which reduces acquisition artifacts, standardizes the intensities of tissues and normalizes the spatial coordinates of brain tissue across MC datasets. Using the standardized datasets, an intuitive set of features based on intensity, spatial location and gradients are extracted and classified using a random forest (RF) classifier to segment the brain tissue class. A series of experiments were conducted to optimize classifier parameters, and to determine segmentation accuracy for standardized and unstandardized (original) data, as a function of scanner vendor, feature type and disease type. The models are trained, tested and validated on 156 image volumes (∼8000 image slices) from two multi-centre, multi-disease datasets, acquired with varying imaging parameters from 30 centres and three scanner vendors. The image datasets, denoted as CAIN and ADNI for vascular and dementia disease, respectively, represent a diverse collection of MC data to test the generalization capabilities of the proposed design. Results demonstrate the importance of standardization for segmentation of MC data, as models trained on standardized data yielded a drastic improvement in brain extraction accuracy compared to the original, unstandardized data (CAIN: DSC = 91% and ADNI: DSC = 86% vs. CAIN: 78% and ADNI: 65%). It was also found that models created from one scanner vendor based on unstandardized data yielded poor segmentation results in data acquired from other scanner vendors, which was improved through standardization. These results demonstrate that to create consistency in segmentations from multi-institutional datasets it is paramount that MC variability be mitigated to improve stability and to ensure generalization of machine learning algorithms for MRI.     

Atomic chain ordering with ultra-long periods: Pb/Si(5 5 7)

As shown previously, Pb on vicinal Si(5 5 7) refacets the surface into a (2 2 3) facet orientation at a Pb coverage of 1.31 ML. This facet formation is electronically stabilized by Fermi nesting and leads to one-dimensional conductance. Electronic correlation seems to be responsible also for the periodic arrangement of atomic Pb chains which decorate the step edges at concentrations exceeding 1.31 ML, up to a concentration of 1.5 ML. Instead of random step decoration, periodicities up to six (2 2 3)-terrace widths (28 lattice constants, 93 Å) have been found. These depend inversely on excess Pb concentration and end at a concentration of 1.52 ML when all steps are decorated with a line density equal to the Si density at steps.These one-dimensional periodicities can be explained assuming that split-off states from surface bands are completely filled by two electrons per Pb atom with corresponding gap opening. This behavior is reminiscent of the formation of charge density waves with tunable wavelengths as a function of excess Pb concentration, and indicates strong electron correlation in this strongly anisotropic 2d system. The alternative, simple band filling within a rigid band model is expected to destabilize the (2 2 3) facet structure upon further adsorption of Pb, which has not been observed.     

Enhanced Slime Mould Algorithm for Multilevel Thresholding Image Segmentation Using Entropy Measures

Image segmentation is a fundamental but essential step in image processing because it dramatically influences posterior image analysis. Multilevel thresholding image segmentation is one of the most popular image segmentation techniques, and many researchers have used meta-heuristic optimization algorithms (MAs) to determine the threshold values. However, MAs have some defects; for example, they are prone to stagnate in local optimal and slow convergence speed. This paper proposes an enhanced slime mould algorithm for global optimization and multilevel thresholding image segmentation, namely ESMA. First, the Levy flight method is used to improve the exploration ability of SMA. Second, quasi opposition-based learning is introduced to enhance the exploitation ability and balance the exploration and exploitation. Then, the superiority of the proposed work ESMA is confirmed concerning the 23 benchmark functions. Afterward, the ESMA is applied in multilevel thresholding image segmentation using minimum cross-entropy as the fitness function. We select eight greyscale images as the benchmark images for testing and compare them with the other classical and state-of-the-art algorithms. Meanwhile, the experimental metrics include the average fitness (mean), standard deviation (Std), peak signal to noise ratio (PSNR), structure similarity index (SSIM), feature similarity index (FSIM), and Wilcoxon rank-sum test, which is utilized to evaluate the quality of segmentation. Experimental results demonstrated that ESMA is superior to other algorithms and can provide higher segmentation accuracy.     

MR image-based measurement of rates of change in volumes of brain structures. Part I: method and validation

A detailed analysis procedure is described for evaluating rates of volumetric change in brain structures based on structural magnetic resonance (MR) images. In this procedure, a series of image processing tools have been employed to address the problems encountered in measuring rates of change based on structural MR images. These tools include an algorithm for intensity non-uniformity correction, a robust algorithm for three-dimensional image registration with sub-voxel precision and an algorithm for brain tissue segmentation. However, a unique feature in the procedure is the use of a fractional volume model that has been developed to provide a quantitative measure for the partial volume effect. With this model, the fractional constituent tissue volumes are evaluated for voxels at the tissue boundary that manifest partial volume effect, thus allowing tissue boundaries be defined at a sub-voxel level and in an automated fashion. Validation studies are presented on key algorithms including segmentation and registration. An overall assessment of the method is provided through the evaluation of the rates of brain atrophy in a group of normal elderly subjects for which the rate of brain atrophy due to normal aging is predictably small. An application of the method is given in Part II where the rates of brain atrophy in various brain regions are studied in relation to normal aging and Alzheimer's disease.     

Evaluation of discrete orthogonal versus polar Stockwell Transform for local multi-resolution texture analysis using brain MRI of multiple sclerosis patients

The Stockwell Transform has the potential to perform multi-resolution texture analysis in magnetic resonance imaging (MRI). However, it is computationally intensive and memory demanding. The polar Stockwell Transform (PST) is rotation-invariant and relatively memory efficient, but still computationally demanding. The new Discrete Orthogonal Stockwell Transform (DOST) appears to have addressed both the computation and storage challenges; however, its utility in localized texture analysis remains unclear. Our goal was to investigate the theory and texture analysis ability of the DOST versus PST using both synthetic and MR images, and explore the relative importance of the associated texture features using a simple classification example based on clinical brain MRI of six multiple sclerosis patients. MRI texture analysis focused on FLAIR images, and the classification used a machine learning algorithm, random forest, that differentiated regions of interest (ROIs) into 2 classes: white matter lesions, and the contralateral normal-appearing white matter (control). Our results showed that the PST features had a greater ability in detecting subtle changes in image structure than the DOST and polar-index DOST (PDOST). Quantitatively, based on 187 lesion and 187 control ROIs, both the PST and the rotation-invariant radial PST performed better in the classification than the DOST and PDOST, where the latter were no better than guessing (p = 0.65 and 0.98). Further analysis using a hierarchical random forest showed that combining MRI signal intensity with the PST or DOST predictions increased the classification performance, with the accuracy, sensitivity, and specificity all improved to >85% in the tests. Collectively, the DOST is less competitive than the PST in localized image texture analysis. The PST features may help with texture-based lesion classification in MS based on clinical brain MRI scans following further verification.     

Edge linking and image segmentation by combining optical and digital methods

We present a hybrid method for segmentation of intensity images, which combines an optical contouring technique and digital algorithms for linking edge points or image segmentation. In a first stage, the digital image to be processed is displayed in a twisted-nematic liquid-crystal display (LCD), which is placed between a polarizer–analyzer pair at 45 deg (instead of 90 deg as occurs in standard LCDs). It is not difficult to demonstrate that the proposed setup produces a resultant image with very pronounced dark contours at middle intensity. After the optical preprocessing, two different digital algorithms are applied: an edge linking algorithm (modified chain code) and a simple thresholding technique for image segmentation. The proposed procedure works well with monochromatic and color images. The method could be useful as a robust technique for segmentation of large images in real-time, which presents potential applications in medical and biological imaging.     

Perpendicular magnetic anisotropy of ultrathin FeCo alloy films on Pd(0 0 1) surface: First principles study

Using the full potential linearized augmented plane wave (FLAPW) method, thickness dependent magnetic anisotropy of ultrathin FeCo alloy films in the range of 1 monolayer (ML) to 5 ML coverage on Pd(0 0 1) surface has been explored. We have found that the FeCo alloy films have close to half metallic state and well-known surface enhancement in thin film magnetism is observed in Fe atom, whereas the Co has rather stable magnetic moment. However, the largest magnetic moment in Fe and Co is found at 1 ML thickness. Interestingly, it has been observed that the interface magnetic moments of Fe and Co are almost the same as those of surface elements. The similar trend exists in orbital magnetic moment. This indicates that the strong hybridization between interface FeCo alloy and Pd gives rise to the large magnetic moment. Theoretically calculated magnetic anisotropy shows that the 1 ML FeCo alloy has in-plane magnetization, but the spin reorientation transition (SRT) from in-plane to perpendicular magnetization is observed above 2 ML thickness with huge magnetic anisotropy energy. The maximum magnetic anisotropy energy for perpendicular magnetization is as large as 0.3 meV/atom at 3 ML film thickness with saturation magnetization of . Besides, the calculated X-ray magnetic circular dichroism (XMCD) has been presented.     

Ultrathin nickel oxide films grown on Ag(0 0 1): a study by XPS, LEIS and LEED intensity analysis

The structure of a nickel oxide film 2 ML thick has been investigated by LEED intensity analysis. The NiO film was prepared by evaporating Ni in presence of O₂ at a pressure in the 10⁻⁶ mbar range. The growth of the oxide film was followed by XPS, LEIS and LEED. In the early stages of deposition, the film shows a (2 × 1) superstructure in LEED. After deposition of 2 ML of NiO, a sharp (1 × 1) LEED pattern is observed. The intensity versus electron energy curves of the LEED spots were measured for this NiO(1 × 1) film and analysed by means of the tensor LEED method. A good level of agreement of the experimental LEED intensities with those calculated for a pseudomorphic NiO(0 0 1) film was obtained. We found that oxygen atoms at the oxide-substrate interface are on-top silver atoms. The interlayer distance in the oxide does not differ significantly from that in bulk NiO(0 0 1), within the accuracy of the analysis. An outward displacement (0.05 ± 0.05 Å) of oxygen atoms with respect to nickel atoms was found at the oxide film surface. The interlayer distance at the silver-nickel oxide interface is 2.43 ± 0.05 Å.