Combining regression trees and radial basis function networks

We describe a method for non-parametric regression which combines regression trees with radial basis function networks. The method is similar to that of Kubat, who was first to suggest such a combination, but has some significant improvements. We demonstrate the features of the new method, compare its performance with other methods on DELVE data sets and apply it to a real world problem involving the classification of soybean plants from digital images.     

Text-independent speaker verification using Minimal Resource Allocation Networks

This paper presents a text-independent speaker verification system based on an online Radial Basis Function (RBF) network referred to as Minimal Resource Allocation Network (MRAN). MRAN is a sequential learning RBF, in which hidden neurons are added or removed as training progresses. LP-derived cepstral coefficients are used as feature vectors during training and verification phases. The performance of MRAN is compared with other well-known RBF and Elliptical Basis Function (EBF) based speaker verification methods in terms of error rates and computational complexity on a series of speaker verification experiments. The experiments use data from 258 speakers from the phonetically balancedcontinuous speech corpus TIMIT. The results show that MRAN produces comparable error rates to other methods with much less computational complexity.     

Effect of polynomial,radial basis,and Pearson VII function kernels in support vector machine algorithm for classification of crayfish

Freshwater crayfish are one of the most important aquatic organisms that play a pivotal role in the aquatic food chain as well as serving as bioindicators for the aquatic ecosystem health assessment. Hemocytes, the blood cells of crustaceans, can be considered stress and health indicators in crayfish, and are used to evaluate the health response. Therefore, total hemocyte cell numbers (THCs) are useful parameters to show the health of crustaceans and serve as stress indicators to decide the quality of the habitat. Since, catching the fish and the other aquatic organisms, and collecting the data for further assessments are time-consuming and frustrating, today, scientists tend to use swift, more sophisticated, and more reliable methods for modeling the ecosystem stressors based on bioindicators. One tool which has attracted the attention of science communities in the last decades is machine learning algorithms that are reliable and accurate methods to solve classification and regression problems. In this study, a support vector machine is carried out as a machine learning algorithm to classify healthy and unhealthy crayfish based on physiological characteristics. To solve the non-linearity problem of the data by transporting data to high-dimensional space, different kernel functions including polynomial (PK), Pearson VII function-based universal (PUK), and radial basis function (RBF) kernels are used and their effect on the performance of the SVM model was evaluated. Both PK and PUK functions performed well in classifying the crayfish. RBF, however, had an adverse impact on the performance of the model. PUK kernel exhibited an outstanding performance (Accuracy = 100%) for the classification of the healthy and unhealthy crayfish.     

Comparative study of SVM methods combined with voxel selection for object category classification on fMRI data

Background

Support vector machine (SVM) has been widely used as accurate and reliable method to decipher brain patterns from functional MRI (fMRI) data. Previous studies have not found a clear benefit for non-linear (polynomial kernel) SVM versus linear one. Here, a more effective non-linear SVM using radial basis function (RBF) kernel is compared with linear SVM. Different from traditional studies which focused either merely on the evaluation of different types of SVM or the voxel selection methods, we aimed to investigate the overall performance of linear and RBF SVM for fMRI classification together with voxel selection schemes on classification accuracy and time-consuming.

Methodology/Principal Findings

Six different voxel selection methods were employed to decide which voxels of fMRI data would be included in SVM classifiers with linear and RBF kernels in classifying 4-category objects. Then the overall performances of voxel selection and classification methods were compared. Results showed that: (1) Voxel selection had an important impact on the classification accuracy of the classifiers: in a relative low dimensional feature space, RBF SVM outperformed linear SVM significantly; in a relative high dimensional space, linear SVM performed better than its counterpart; (2) Considering the classification accuracy and time-consuming holistically, linear SVM with relative more voxels as features and RBF SVM with small set of voxels (after PCA) could achieve the better accuracy and cost shorter time.

Conclusions/Significance

The present work provides the first empirical result of linear and RBF SVM in classification of fMRI data, combined with voxel selection methods. Based on the findings, if only classification accuracy was concerned, RBF SVM with appropriate small voxels and linear SVM with relative more voxels were two suggested solutions; if users concerned more about the computational time, RBF SVM with relative small set of voxels when part of the principal components were kept as features was a better choice.     

Insulin induces the correlation between renal blood flow and glomerular filtration rate in diabetes: implications for mechanisms causing hyperfiltration

Glomerular filtration rate (GFR) and renal blood flow (RBF) are normally kept constant via renal autoregulation. However, early diabetes results in increased GFR and the potential mechanisms are debated. Tubuloglomerular feedback (TGF) inactivation, with concomitantly increased RBF, is proposed but challenged by the finding of glomerular hyperfiltration in diabetic adenosine A(1) receptor-deficient mice, which lack TGF. Furthermore, we consistently find elevated GFR in diabetes with only minor changes in RBF. This may relate to the use of a lower streptozotocin dose, which produces a degree of hyperglycemia, which is manageable without supplemental suboptimal insulin administration, as has been used by other investigators. Therefore, we examined the relationship between RBF and GFR in diabetic rats with (diabetes + insulin) and without suboptimal insulin administration (untreated diabetes). As insulin can affect nitric oxide (NO) release, the role of NO was also investigated. GFR, RBF, and glomerular filtration pressures were measured. Dynamic RBF autoregulation was examined by transfer function analysis between arterial pressure and RBF. Both diabetic groups had increased GFR (+60-67%) and RBF (+20-23%) compared with controls. However, only the diabetes + insulin group displayed a correlation between GFR and RBF (R(2) = 0.81, P < 0.0001). Net filtration pressure was increased in untreated diabetes compared with both other groups. The difference between untreated and insulin-treated diabetic rats disappeared after administering N(ω)-nitro-l-arginine methyl ester to inhibit NO synthase and subsequent NO release. In conclusion, mechanisms causing diabetes-induced glomerular hyperfiltration are animal model-dependent. Supplemental insulin administration results in a RBF-dependent mechanism, whereas elevated GFR in untreated diabetes is mediated primarily by a tubular event. Insulin-induced NO release partially contributes to these differences.     

支持向量机与神经网络的关系研究

支持向量机是一种基于统计学习理论的新颖的机器学习方法,由于其出色的学习性能,该技术已成为当前国际机器学习界的研究热点,该方法已经广泛用于解决分类和回归问题．本文将结构风险函数应用于径向基函数网络学习中,同时讨论了支持向量回归模型和径向基函数网络之间的关系．仿真实例表明所给算法提高了径向基函数网络的泛化性能．     

Complex-valued minimal resource allocation network for nonlinear signal processing

This paper presents a sequential learning algorithm and evaluates its performance on complex valued signal processing problems. The algorithm is referred to as Complex Minimal Resource Allocation Network (CMRAN) algorithm and it is an extension of the MRAN algorithm originally developed for online learning in real valued RBF networks. CMRAN has the ability to grow and prune the (complex) RBF network's hidden neurons to ensure a parsimonious network structure. The performance of the learning algorithm is illustrated using two applications from signal processing of communication systems. The first application considers identification of a nonlinear complex channel. The second application considers the use of CMRAN to QAM digital channel equalization problems. Simulation results presented clearly show that CMRAN is very effective in modeling and equalization with performance achieved often being superior to that of some of the well known methods.     

Binding effect of polychlorinated compounds and environmental carcinogens on rice bran fiber

To accelerate the fecal excretion of polycyclic biphenyl (PCB), polychlorinated dibenzofurans (PCDFs), polychlorinated-p-dioxines (PCDDs) and various mutagens and carcinogens, their binding effect on rice bran fiber (RBF) was investigated for nine heterocyclic amines, six nitroarenes, 4-nitroquinoline-N-oxide, benzo[a]pyrene, furylfuramide, two kinds of flavonoid compounds and formaldehyde and ascorbic acid. PCBs, PCDFs and PCDDs suspended in nonane were incubated with RBF (10 mg/ml) at 37 degrees C and after centrifugation, unbound chemicals in the supernatant were analyzed by high-performance liquid chromatography (HPLC) and gas chromatography (GC). The binding effects on RBF were enhanced more than other dietary fibers (DFs), which were tested including corn, wheat bran, spinach, Hijiki (a kind of seaweed), sweet potatoes and burdock fibers. It was found that the binding effects were related to lignin contents. Binding of 3-amino-1(or 1,4)-dimethyl-5H-pyrido[4,3-b]indole (Trp-p-1 and Trp-p-2), food-derived carcinogens and 1-nitropyrene (1-NP), suspended in methanol, to RBF occurred within 10 min of incubation at 37 degrees C at pH 5-7, and decreased below pH 4; binding of food-derived carcinogens was pH dependent. The binding effects to RBF and pulp lignin were obtained at ratio of over 90%, while corn fiber and cellulose were at ratios of 4-30%. Polycyclic aromatic compounds were related to the number of rings, showing high binding effects to chemical structures with triple rings. Binding of 1-NP and PCB to RBF was not influenced in any aerobic and anaerobic bacterial cultures. It was also found that RBF was capable of binding even conjugates containing mutagens such as glucuronides and sulfates, as well as metabolites in urine. It was suggested, therefore, that mutagens and carcinogens were available for the fecal excretion of residual chemicals and their metabolites, and also for the fecal excretion of PCBs, PCDFs and related compound residues in patients of Yusho disease, who suffered food poisoning due to rice oil contaminated with PCB in Japan.     

Improved Classification of Lung Cancer Using Radial Basis Function Neural Network with Affine Transforms of Voss Representation

Lung cancer is one of the diseases responsible for a large number of cancer related death cases worldwide. The recommended standard for screening and early detection of lung cancer is the low dose computed tomography. However, many patients diagnosed die within one year, which makes it essential to find alternative approaches for screening and early detection of lung cancer. We present computational methods that can be implemented in a functional multi-genomic system for classification, screening and early detection of lung cancer victims. Samples of top ten biomarker genes previously reported to have the highest frequency of lung cancer mutations and sequences of normal biomarker genes were respectively collected from the COSMIC and NCBI databases to validate the computational methods. Experiments were performed based on the combinations of Z-curve and tetrahedron affine transforms, Histogram of Oriented Gradient (HOG), Multilayer perceptron and Gaussian Radial Basis Function (RBF) neural networks to obtain an appropriate combination of computational methods to achieve improved classification of lung cancer biomarker genes. Results show that a combination of affine transforms of Voss representation, HOG genomic features and Gaussian RBF neural network perceptibly improves classification accuracy, specificity and sensitivity of lung cancer biomarker genes as well as achieving low mean square error.     

基于辅助环境变量的土壤有机碳空间插值——以黄土丘陵区小流域为例

利用普通克里格法(OK)、反距离加权法(IDW)、径向基函数法(RBF)、基于土地利用类型修正的普通克里格法(OK_LU)4种插值方法,对黄土丘陵羊圈沟小流域的土壤有机碳含量进行空间插值。预测结果的准确性通过Pearson相关系数(R),平均绝对误差(MAE),均方根误差(RMSE),准确度(AC)来评价。研究结果表明:(1)在前3种常规空间插值方法中,OK对刻画区域土壤有机碳的空间分布趋势效果最佳,其预测MAE值和RMSE值均为最小,Pearson相关系数(R)和准确度(AC)最大,说明其预测结果的准确性最好、预测的极端误差也最小;其次为RBF;IDW预测的效果最差。(2)OK_LU在空间特征表达方面能够更好地反映复杂地形区的局部变异,其插值结果的精度相比OK有一定程度的提高,其平均绝对误差(MAE)从0.900%降到了0.567%,均方根误差(RMSE)从1.101%降到了0.777%,Pearson相关系数(R)从0.4026提高到0.5589,准确度(AC)从0.9081提高到0.9505。综合比较,在黄土丘陵地区,OK_LU能使插值结果的精度有较大提高,是土壤有机碳空间制图的有效途径。     

Systematic benchmarking of microarray data classification: assessing the role of non-linearity and dimensionality reduction

MOTIVATION: Microarrays are capable of determining the expression levels of thousands of genes simultaneously. In combination with classification methods, this technology can be useful to support clinical management decisions for individual patients, e.g. in oncology. The aim of this paper is to systematically benchmark the role of non-linear versus linear techniques and dimensionality reduction methods. RESULTS: A systematic benchmarking study is performed by comparing linear versions of standard classification and dimensionality reduction techniques with their non-linear versions based on non-linear kernel functions with a radial basis function (RBF) kernel. A total of 9 binary cancer classification problems, derived from 7 publicly available microarray datasets, and 20 randomizations of each problem are examined. CONCLUSIONS: Three main conclusions can be formulated based on the performances on independent test sets. (1) When performing classification with least squares support vector machines (LS-SVMs) (without dimensionality reduction), RBF kernels can be used without risking too much overfitting. The results obtained with well-tuned RBF kernels are never worse and sometimes even statistically significantly better compared to results obtained with a linear kernel in terms of test set receiver operating characteristic and test set accuracy performances. (2) Even for classification with linear classifiers like LS-SVM with linear kernel, using regularization is very important. (3) When performing kernel principal component analysis (kernel PCA) before classification, using an RBF kernel for kernel PCA tends to result in overfitting, especially when using supervised feature selection. It has been observed that an optimal selection of a large number of features is often an indication for overfitting. Kernel PCA with linear kernel gives better results.     

海洋浮游植物丰度的空间插值优化

在探索性空间数据分析(Exploratory spatial data analysis)和数据转化的基础上,利用反距离加权(Inverse distance weighting,IDW)、径向基函数(Radial basis functions,RBF)、普通克里格(Ordinary Kriging,OK),3种插值方法,对2003年8月获得的珠江口浮游植物丰度数据进行插值运算,并对插值准确度进行交叉验证。结果显示,珠江口浮游植物丰度数据具有离散性大、存在极大和极小值、呈正偏分布等特点。而对数转化能大大减小数据的离散性和不对称性,有效消除插值结果图中各类插值噪音。交叉验证显示,插值精确度OK最高,RBF次之,IDW最低。观察插值结果等值面图,发现3种方法均能较客观地模拟出浮游植物丰度的总体分布趋势,在对局部趋势的模拟上,OK的表现最好。综合评定,OK为最适合珠江口浮游植物丰度数据的插值方法。半变异模型的选择对OK的插值结果影响不明显。在四种半变异模型中,圆形模型的拟合效果最好。     

Marker-assisted prediction of non-additive genetic values

It has become increasingly clear from systems biology arguments that interaction and non-linearity play an important role in genetic regulation of phenotypic variation for complex traits. Marker-assisted prediction of genetic values assuming additive gene action has been widely investigated because of its relevance in artificial selection. On the other hand, it has been less well-studied when non-additive effects hold. Here, we explored a nonparametric model, radial basis function (RBF) regression, for predicting quantitative traits under different gene action modes (additivity, dominance and epistasis). Using simulation, it was found that RBF had better ability (higher predictive correlations and lower predictive mean square errors) of predicting merit of individuals in future generations in the presence of non-additive effects than a linear additive model, the Bayesian Lasso. This was true for populations undergoing either directional or random selection over several generations. Under additive gene action, RBF was slightly worse than the Bayesian Lasso. While prediction of genetic values under additive gene action is well handled by a variety of parametric models, nonparametric RBF regression is a useful counterpart for dealing with situations where non-additive gene action is suspected, and it is robust irrespective of mode of gene action.     

A comparative study of cell classifiers for image-based high-throughput screening

Background

Millions of cells are present in thousands of images created in high-throughput screening (HTS). Biologists could classify each of these cells into a phenotype by visual inspection. But in the presence of millions of cells this visual classification task becomes infeasible. Biologists train classification models on a few thousand visually classified example cells and iteratively improve the training data by visual inspection of the important misclassified phenotypes. Classification methods differ in performance and performance evaluation time. We present a comparative study of computational performance of gentle boosting, joint boosting CellProfiler Analyst (CPA), support vector machines (linear and radial basis function) and linear discriminant analysis (LDA) on two data sets of HT29 and HeLa cancer cells.

Results

For the HT29 data set we find that gentle boosting, SVM (linear) and SVM (RBF) are close in performance but SVM (linear) is faster than gentle boosting and SVM (RBF). For the HT29 data set the average performance difference between SVM (RBF) and SVM (linear) is 0.42 %. For the HeLa data set we find that SVM (RBF) outperforms other classification methods and is on average 1.41 % better in performance than SVM (linear).

Conclusions

Our study proposes SVM (linear) for iterative improvement of the training data and SVM (RBF) for the final classifier to classify all unlabeled cells in the whole data set.

Electronic supplementary material

The online version of this article (doi:10.1186/1471-2105-15-342) contains supplementary material, which is available to authorized users.     

Modeling and optimization of microbial hyaluronic acid production by Streptococcus zooepidemicus using radial basis function neural network coupling quantum‐behaved particle swarm optimization algorithm

Hyaluronic acid (HA) is a natural biopolymer with unique physiochemical and biological properties and finds a wide range of applications in biomedical and cosmetic fields. It is important to increase HA production to meet the increasing HA market demand. This work is aimed to model and optimize the amino acids addition to enhance HA production of Streptococcus zooepidemicus with radial basis function (RBF) neural network coupling quantum‐behaved particle swarm optimization (QPSO) algorithm. In the RBF‐QPSO approach, RBF neural network is used as a bioprocess modeling tool and QPSO algorithm is applied to conduct the optimization with the established RBF neural network black model as the objective function. The predicted maximum HA yield was 6.92 g/L under the following conditions: arginine 0.062 g/L, cysteine 0.036 g/L, and lysine 0.043 g/L. The optimal amino acids addition allowed HA yield increased from 5.0 g/L of the control to 6.7 g/L in the validation experiments. Moreover, the modeling and optimization capacity of the RBF‐QPSO approach was compared with that of response surface methodology (RSM). It was indicated that the RBF‐QPSO approach gave a slightly better modeling and optimization result compared with RSM. The developed RBF‐QPSO approach in this work may be helpful for the modeling and optimization of the other multivariable, nonlinear, time‐variant bioprocesses. © 2009 American Institute of Chemical Engineers Biotechnol. Prog., 2009     

Local thermodilution: a reliable technique for estimating renal blood flow in the rabbit

1. A thermistor probe designed for determination of renal blood flow in rabbits, consisted of a fast-responding bead thermistor and an injection port which was also used to measure renal venous pressure between injections. 2. By an in vitro calibration system, actual measured flow (Qa) correlates well with the thermodilution calculated flow (Qc), where Qc = 0.99 Qa + 4.9 (r = 0.97, n = 42). 3. The renal blood flow (RBF) as determined by the thermodilution technique in 3 control groups was 53 +/- 3 (8), 60 +/- 6 (8), and 62 +/- 3 (3) ml/min/kidney or about 9% of the cardiac output. 4. Hypovolemia (-10%) reduced RBF by 19% from the control value, whereas, hypervolemia (+10%) did not alter RBF. 5. Smoke-induced apnea resulted in hypertension (+30%) and bradycardia (-39%), and was associated with changes in RBF (-55%) and renal vascular resistance (+183%). 6. We conclude that the local thermodilution technique is a relatively easy and reliable method for estimating RBF in rabbits.     

Long-term control of renal blood flow: what is the role of the renal nerves?

We have developed a system for long-term continuous monitoring of cardiovascular parameters in rabbits living in their home cage to assess what role renal sympathetic nerve activity (RSNA) has in regulating renal blood flow (RBF) in daily life. Blood pressure, heart rate, locomotor activity, RSNA, and RBF were recorded continuously for 4 wk. Beginning 4-5 days after surgery a circadian rhythm, dependent on feeding time, was observed. When averaged over all days RBF to the innervated and denervated kidneys was not significantly different. However, control of RBF around these mean levels was dependent on the presence of the renal sympathetic nerves. In particular we observed episodic elevations in heart rate and other parameters associated with activity. In the denervated kidney, during these episodic elevations, the increase in renal resistance was closely related to the increase in arterial pressure. In the innervated kidney the renal resistance response was significantly more variable, indicating an interaction of the sympathetic nervous system. These results indicate that whereas overall levels of RSNA do not set the mean level of RBF the renal vasculature is sensitive to episodic increases in sympathetic nerve activity.