Construction of embedded fMRI resting-state functional connectivity networks using manifold learning

We construct embedded functional connectivity networks (FCN) from benchmark resting-state functional magnetic resonance imaging (rsfMRI) data acquired from patients with schizophrenia and healthy controls based on linear and nonlinear manifold learning algorithms, namely, Multidimensional Scaling, Isometric Feature Mapping, Diffusion Maps, Locally Linear Embedding and kernel PCA. Furthermore, based on key global graph-theoretic properties of the embedded FCN, we compare their classification potential using machine learning. We also assess the performance of two metrics that are widely used for the construction of FCN from fMRI, namely the Euclidean distance and the cross correlation metric. We show that diffusion maps with the cross correlation metric outperform the other combinations.     

基于核局部线性嵌入的基因表达谱数据分类

针对局部线性嵌入算法（LocalLinearEmbedding,LLE）利用试凑法寻找近邻数耗时的缺陷性,提出一种增强的核局部线性嵌入算法（EnhancedKernelLocalLinearEmbedding,EKLLE）自动为样本分配邻域;该算法以高斯核函数为核心改进标准LLE距离度量准则,结合样本的类别信息,无需人工干预自动为样本设置不同的近邻数,克服了试凑法获得最优结果时需要大量时间;最后在各样本近邻数不相同的情况下对数据进行维数简约及待测样本分类。EKLLE算法有效地将高维基因表达谱数据映射到低维本质空间中,解决了传统LLE算法不能很好地处理合噪声或者稀疏数据的缺点。通过对比其他肿瘤样本分类实验,验证本文方法的实时性和精确性。     

基于F3Net显著性目标检测的蝴蝶图像前背景自动分割

【目的】具有复杂背景的蝴蝶图像前背景分割难度大。本研究旨在探索基于深度学习显著性目标检测的蝴蝶图像自动分割方法。【方法】应用DUTS-TR数据集训练F³Net显著性目标检测算法构建前背景预测模型,然后将模型用于具有复杂背景的蝴蝶图像数据集实现蝴蝶前背景自动分割。在此基础上,采用迁移学习方法,保持ResNet骨架不变,利用蝴蝶图像及其前景蒙板数据,使用交叉特征模块、级联反馈解码器和像素感知损失方法重新训练优化模型参数,得到更优的自动分割模型。同时,将其他5种基于深度学习显著性检测算法也用于自动分割,并比较了这些算法和F³Net算法的性能。【结果】所有算法均获得了很好的蝴蝶图像前背景分割效果,其中,F³Net是更优的算法,其7个指标S测度、E测度、F测度、平均绝对误差(MAE)、精度、召回率和平均IoU值分别为0.940, 0.945, 0.938, 0.024, 0.929,0.978和0.909。迁移学习则进一步提升了F³Net的上述指标值,分别为0.961, 0.964, 0.963, 0.013, 0.965, 0.967和0.938。【结论】研究结果证明结合迁移学习的F³Net算法是其中最优的分割方法。本研究提出的方法可用于野外调查中拍摄的昆虫图像的自动分割,并拓展了显著性目标检测方法的应用范围。     

基于对象特征的山东省丘陵地区多时相遥感土地覆被自动分类

对于基于像元的土地覆被分类来说,植被的分类是难点。使用多时相面向对象分类方法可以较好的解决这个问题。以山东省烟台市丘陵地区为研究区,采用Landsat TM(Landsat Thematic Mapper remotely sensed imagery)、DEM(Digital Elevation Model)、坡度、坡位、坡向等多种数据,利用基于对象特征的多时相分类方法对研究区进行土地覆盖自动分类。首先对影像进行多尺度分割并检验分割结果选取合适的分割尺度,然后分析对象的光谱、纹理、形状特征。根据各类地物的光谱特征、地理相关性、形状、空间分布等特征,明确类别之间的差异。建立决策树使用隶属度函数进行模糊分类,借助支持向量机提高分类精度。研究结果表明,通过使用多时相影像采用面向对象分类方法,相对于传统的基于像素的分类可以明显提高分类精度,尤其是解决了乔灌草的区分问题。     

Segmentation of the Clustered Cells with Optimized Boundary Detection in Negative Phase Contrast Images

Cell image segmentation plays a central role in numerous biology studies and clinical applications. As a result, the development of cell image segmentation algorithms with high robustness and accuracy is attracting more and more attention. In this study, an automated cell image segmentation algorithm is developed to get improved cell image segmentation with respect to cell boundary detection and segmentation of the clustered cells for all cells in the field of view in negative phase contrast images. A new method which combines the thresholding method and edge based active contour method was proposed to optimize cell boundary detection. In order to segment clustered cells, the geographic peaks of cell light intensity were utilized to detect numbers and locations of the clustered cells. In this paper, the working principles of the algorithms are described. The influence of parameters in cell boundary detection and the selection of the threshold value on the final segmentation results are investigated. At last, the proposed algorithm is applied to the negative phase contrast images from different experiments. The performance of the proposed method is evaluated. Results show that the proposed method can achieve optimized cell boundary detection and highly accurate segmentation for clustered cells.     

Color image analysis for quantifying renal tumor angiogenesis

OBJECTIVE: To segment and quantify microvessels in renal tumor angiogenesis based on a color image analysis method and to improve the accuracy and reproducibility of quantifying microvessel density. STUDY DESIGN: The segmentation task was based on a supervised learning scheme. First, 12 color features (RGB, HSI, I1I2I3 and L*a*b*) were extracted from a training set. The feature selection procedure selected I2L*S features as the best color feature vector. Then we segmented microvessels using the discriminant function made using the minimum error rate classification rule of Bayesian decision theory. In the quantification step, after applying a connected component-labeling algorithm, microvessels with discontinuities were connected and touching microvessels separated. We tested the proposed method on 23 images. RESULTS: The results were evaluated by comparing them with manual quantification of the same images. The comparison revealed that our computerized microvessel counting correlated highly with manual counting by an expert (r = 0.95754). The association between the number of microvessels after the initial segmentation and manual quantification was also assessed using Pearson's correlation coefficient (r = 0.71187). The results indicate that our method is better than conventional computerized image analysis methods. CONCLUSION: Our method correlated highly with quantification by an expert and could become a way to improve the accuracy, feasibility and reproducibility of quantifying microvessel density. We anticipate that it will become a useful diagnostic tool for angiogenesis studies.     

Two-Stage Feature Selection of Voice Parameters for Early Alzheimer's Disease Prediction

Background

The goal of this work is to develop a non-invasive method in order to help detecting Alzheimer's disease in its early stages, by implementing voice analysis techniques based on machine learning algorithms.

Comparing Artificial Neural Networks,General Linear Models and Support Vector Machines in Building Predictive Models for Small Interfering RNAs

Background

Exogenous short interfering RNAs (siRNAs) induce a gene knockdown effect in cells by interacting with naturally occurring RNA processing machinery. However not all siRNAs induce this effect equally. Several heterogeneous kinds of machine learning techniques and feature sets have been applied to modeling siRNAs and their abilities to induce knockdown. There is some growing agreement to which techniques produce maximally predictive models and yet there is little consensus for methods to compare among predictive models. Also, there are few comparative studies that address what the effect of choosing learning technique, feature set or cross validation approach has on finding and discriminating among predictive models.

Principal Findings

Three learning techniques were used to develop predictive models for effective siRNA sequences including Artificial Neural Networks (ANNs), General Linear Models (GLMs) and Support Vector Machines (SVMs). Five feature mapping methods were also used to generate models of siRNA activities. The 2 factors of learning technique and feature mapping were evaluated by complete 3×5 factorial ANOVA. Overall, both learning techniques and feature mapping contributed significantly to the observed variance in predictive models, but to differing degrees for precision and accuracy as well as across different kinds and levels of model cross-validation.

Conclusions

The methods presented here provide a robust statistical framework to compare among models developed under distinct learning techniques and feature sets for siRNAs. Further comparisons among current or future modeling approaches should apply these or other suitable statistically equivalent methods to critically evaluate the performance of proposed models. ANN and GLM techniques tend to be more sensitive to the inclusion of noisy features, but the SVM technique is more robust under large numbers of features for measures of model precision and accuracy. Features found to result in maximally predictive models are not consistent across learning techniques, suggesting care should be taken in the interpretation of feature relevance. In the models developed here, there are statistically differentiable combinations of learning techniques and feature mapping methods where the SVM technique under a specific combination of features significantly outperforms all the best combinations of features within the ANN and GLM techniques.     

SLLE for predicting membrane protein types

Introduction of the concept of pseudo amino acid composition (PROTEINS: Structure, Function, and Genetics 43 (2001) 246; Erratum: ibid. 44 (2001) 60) has made it possible to incorporate a considerable amount of sequence-order effects by representing a protein sample in terms of a set of discrete numbers, and hence can significantly enhance the prediction quality of membrane protein type. As a continuous effort along such a line, the Supervised Locally Linear Embedding (SLLE) technique for nonlinear dimensionality reduction is introduced (Science 22 (2000) 2323). The advantage of using SLLE is that it can reduce the operational space by extracting the essential features from the high-dimensional pseudo amino acid composition space, and that the cluster-tolerant capacity can be increased accordingly. As a consequence by combining these two approaches, high success rates have been observed during the tests of self-consistency, jackknife and independent data set, respectively, by using the simplest nearest neighbour classifier. The current approach represents a new strategy to deal with the problems of protein attribute prediction, and hence may become a useful vehicle in the area of bioinformatics and proteomics.     

A Systematic Comparison of Supervised Classifiers

Pattern recognition has been employed in a myriad of industrial, commercial and academic applications. Many techniques have been devised to tackle such a diversity of applications. Despite the long tradition of pattern recognition research, there is no technique that yields the best classification in all scenarios. Therefore, as many techniques as possible should be considered in high accuracy applications. Typical related works either focus on the performance of a given algorithm or compare various classification methods. In many occasions, however, researchers who are not experts in the field of machine learning have to deal with practical classification tasks without an in-depth knowledge about the underlying parameters. Actually, the adequate choice of classifiers and parameters in such practical circumstances constitutes a long-standing problem and is one of the subjects of the current paper. We carried out a performance study of nine well-known classifiers implemented in the Weka framework and compared the influence of the parameter configurations on the accuracy. The default configuration of parameters in Weka was found to provide near optimal performance for most cases, not including methods such as the support vector machine (SVM). In addition, the k-nearest neighbor method frequently allowed the best accuracy. In certain conditions, it was possible to improve the quality of SVM by more than 20% with respect to their default parameter configuration.     

Stroke Treatment Prediction Using Features Selection Methods and Machine Learning Classifiers

ObjectivesFeature selection in data sets is an important task allowing to alleviate various machine learning and data mining issues. The main objectives of a feature selection method consist on building simpler and more understandable classifier models in order to improve the data mining and processing performances. Therefore, a comparative evaluation of the Chi-square method, recursive feature elimination method, and tree-based method (using Random Forest) used on the three common machine learning methods (K-Nearest Neighbor, naïve Bayesian classifier and decision tree classifier) are performed to select the most relevant primitives from a large set of attributes. Furthermore, determining the most suitable couple (i.e., feature selection method-machine learning method) that provides the best performance is performed.Materials and methodsIn this paper, an overview of the most common feature selection techniques is first provided: the Chi-Square method, the Recursive Feature Elimination method (RFE) and the tree-based method (using Random Forest). A comparative evaluation of the improvement (brought by such feature selection methods) to the three common machine learning methods (K- Nearest Neighbor, naïve Bayesian classifier and decision tree classifier) are performed. For evaluation purposes, the following measures: micro-F1, accuracy and root mean square error are used on the stroke disease data set.ResultsThe obtained results show that the proposed approach (i.e., Tree Based Method using Random Forest, TBM-RF, decision tree classifier, DTC) provides accuracy higher than 85%, F1-score higher than 88%, thus, better than the KNN and NB using the Chi-Square, RFE and TBM-RF methods.ConclusionThis study shows that the couple - Tree Based Method using Random Forest (TBM-RF) decision tree classifier successfully and efficiently contributes to find the most relevant features and to predict and classify patient suffering of stroke disease.”     

Liver vessel segmentation based on extreme learning machine

Liver-vessel segmentation plays an important role in vessel structure analysis for liver surgical planning. This paper presents a liver-vessel segmentation method based on extreme learning machine (ELM). Firstly, an anisotropic filter is used to remove noise while preserving vessel boundaries from the original computer tomography (CT) images. Then, based on the knowledge of prior shapes and geometrical structures, three classical vessel filters including Sato, Frangi and offset medialness filters together with the strain energy filter are used to extract vessel structure features. Finally, the ELM is applied to segment liver vessels from background voxels. Experimental results show that the proposed method can effectively segment liver vessels from abdominal CT images, and achieves good accuracy, sensitivity and specificity.     

不同滤波方法对超声图像分割的影响

采用各向异性滤波方法以及Gabor滤波方法对乳腺肿瘤超声图像进行处理,再使用snake方法以及levelset方法在设置相同参数的条件下,对过滤的图像分别进行分割。试验结果表明,不同的滤波方法在分害4目标图像的收敛度,边缘圆滑度及整体轮廓提取效果都有较大影响,为分割图像选取适当的滤波器提供了参考。     

Accuracy of protein flexibility predictions

Protein structural flexibility is important for catalysis, binding, and allostery. Flexibility has been predicted from amino acid sequence with a sliding window averaging technique and applied primarily to epitope search. New prediction parameters were derived from 92 refined protein structures in an unbiased selection of the Protein Data Bank by developing further the method of Karplus and Schulz (Naturwissenschaften 72:212–213, 1985). The accuracy of four flexibility prediction techniques was studied by comparing atomic temperature factors of known three-dimensional protein structures to predictions by using correlation coefficients. The size of the prediction window was optimized for each method. Predictions made with our new parameters, using an optimized window size of 9 residues in the prediction window, were giving the best results. The difference from another previously used technique was small, whereas two other methods were much poorer. Applicability of the predictions was also tested by searching for known epitopes from amino acid sequences. The best techniques predicted correctly 20 of 31 continuous epitopes in seven proteins. Flexibility parameters have previously been used for calculating protein average flexibility indices which are inversely correlated to protein stability. Indices with the new parameters showed better correlation to protein stability than those used previously; furthermore they had relationship even when the old parameters failed. © 1994 Wiley-Liss, Inc.     

3-Dimensional Diffusion Tensor Imaging (DTI) Atlas of the Rat Brain

Anatomical atlases play an important role in the analysis of neuroimaging data in rodent neuroimaging studies. Having a high resolution, detailed atlas not only can expand understanding of rodent brain anatomy, but also enables automatic segmentation of new images, thus greatly increasing the efficiency of future analysis when applied to new data. These atlases can be used to analyze new scans of individual cases using a variety of automated segmentation methods. This project seeks to develop a set of detailed 3D anatomical atlases of the brain at postnatal day 5 (P5), 14 (P14), and adults (P72) in Sprague-Dawley rats. Our methods consisted of first creating a template image based on fixed scans of control rats, then manually segmenting various individual brain regions on the template. Using itk-SNAP software, subcortical and cortical regions, including both white matter and gray matter structures, were manually segmented in the axial, sagittal, and coronal planes. The P5, P14, and P72 atlases had 39, 45, and 29 regions segmented, respectively. These atlases have been made available to the broader research community.     

Investigating the efficacy of nonlinear dimensionality reduction schemes in classifying gene and protein expression studies

The recent explosion in procurement and availability of high-dimensional gene- and protein-expression profile datasets for cancer diagnostics has necessitated the development of sophisticated machine learning tools with which to analyze them. A major limitation in the ability to accurate classify these high-dimensional datasets stems from the 'curse of dimensionality', occurring in situations where the number of genes or peptides significantly exceeds the total number of patient samples. Previous attempts at dealing with this issue have mostly centered on the use of a dimensionality reduction (DR) scheme, Principal Component Analysis (PCA), to obtain a low-dimensional projection of the high-dimensional data. However, linear PCA and other linear DR methods, which rely on Euclidean distances to estimate object similarity, do not account for the inherent underlying nonlinear structure associated with most biomedical data. The motivation behind this work is to identify the appropriate DR methods for analysis of high-dimensional gene- and protein-expression studies. Towards this end, we empirically and rigorously compare three nonlinear (Isomap, Locally Linear Embedding, Laplacian Eigenmaps) and three linear DR schemes (PCA, Linear Discriminant Analysis, Multidimensional Scaling) with the intent of determining a reduced subspace representation in which the individual object classes are more easily discriminable.     

Automatic identification of the number of food items in a meal using clustering techniques based on the monitoring of swallowing and chewing

The number of distinct foods consumed in a meal is of significant clinical concern in the study of obesity and other eating disorders. This paper proposes the use of information contained in chewing and swallowing sequences for meal segmentation by food types. Data collected from experiments of 17 volunteers were analyzed using two different clustering techniques. First, an unsupervised clustering technique, Affinity Propagation (AP), was used to automatically identify the number of segments within a meal. Second, performance of the unsupervised AP method was compared to a supervised learning approach based on Agglomerative Hierarchical Clustering (AHC). While the AP method was able to obtain 90% accuracy in predicting the number of food items, the AHC achieved an accuracy >95%. Experimental results suggest that the proposed models of automatic meal segmentation may be utilized as part of an integral application for objective Monitoring of Ingestive Behavior in free living conditions.     

Heuristic Bayesian Segmentation for Discovery of Coexpressed Genes within Genomic Regions

Segmentation aims to separate homogeneous areas from the sequential data, and plays a central role in data mining. It has applications ranging from finance to molecular biology, where bioinformatics tasks such as genome data analysis are active application fields. In this paper, we present a novel application of segmentation in locating genomic regions with coexpressed genes. We aim at automated discovery of such regions without requirement for user-given parameters. In order to perform the segmentation within a reasonable time, we use heuristics. Most of the heuristic segmentation algorithms require some decision on the number of segments. This is usually accomplished by using asymptotic model selection methods like the Bayesian information criterion. Such methods are based on some simplification, which can limit their usage. In this paper, we propose a Bayesian model selection to choose the most proper result from heuristic segmentation. Our Bayesian model presents a simple prior for the segmentation solutions with various segment numbers and a modified Dirichlet prior for modeling multinomial data. We show with various artificial data sets in our benchmark system that our model selection criterion has the best overall performance. The application of our method in yeast cell-cycle gene expression data reveals potential active and passive regions of the genome.     

A Multi-Label Learning Based Kernel Automatic Recommendation Method for Support Vector Machine

Choosing an appropriate kernel is very important and critical when classifying a new problem with Support Vector Machine. So far, more attention has been paid on constructing new kernels and choosing suitable parameter values for a specific kernel function, but less on kernel selection. Furthermore, most of current kernel selection methods focus on seeking a best kernel with the highest classification accuracy via cross-validation, they are time consuming and ignore the differences among the number of support vectors and the CPU time of SVM with different kernels. Considering the tradeoff between classification success ratio and CPU time, there may be multiple kernel functions performing equally well on the same classification problem. Aiming to automatically select those appropriate kernel functions for a given data set, we propose a multi-label learning based kernel recommendation method built on the data characteristics. For each data set, the meta-knowledge data base is first created by extracting the feature vector of data characteristics and identifying the corresponding applicable kernel set. Then the kernel recommendation model is constructed on the generated meta-knowledge data base with the multi-label classification method. Finally, the appropriate kernel functions are recommended to a new data set by the recommendation model according to the characteristics of the new data set. Extensive experiments over 132 UCI benchmark data sets, with five different types of data set characteristics, eleven typical kernels (Linear, Polynomial, Radial Basis Function, Sigmoidal function, Laplace, Multiquadric, Rational Quadratic, Spherical, Spline, Wave and Circular), and five multi-label classification methods demonstrate that, compared with the existing kernel selection methods and the most widely used RBF kernel function, SVM with the kernel function recommended by our proposed method achieved the highest classification performance.     

A Comparative Approach for MI-Based EEG Signals Classification Using Energy,Power and Entropy

ObjectiveThe initial principal task of a Brain-Computer Interfacing (BCI) research is to extract the best feature set from a raw EEG (Electroencephalogram) signal so that it can be used for the classification of two or multiple different events. The main goal of the paper is to develop a comparative analysis among different feature extraction techniques and classification algorithms.Materials and methodsIn this present investigation, four different methodologies have been adopted to classify the recorded MI (motor imagery) EEG signal, and their comparative study has been reported. Haar Wavelet Energy (HWE), Band Power, Cross-correlation, and Spectral Entropy (SE) based Cross-correlation feature extraction techniques have been considered to obtain the necessary features set from the raw EEG signals. Four different machine learning algorithms, viz. LDA (Linear Discriminant Analysis), QDA (Quadratic Discriminant Analysis), Naïve Bayes, and Decision Tree, have been used to classify the features.ResultsThe best average classification accuracies are 92.50%, 93.12%, 72.26%, and 98.71% using the four methods. Further, these results have been compared with some recent existing methods.ConclusionThe comparative results indicate a significant accuracy level performance improvement of the proposed methods with respect to the existing one. Hence, this presented work can guide to select the best feature extraction method and the classifier algorithm for MI-based EEG signals.