Meihong Wu

Improving tRNAscan-SE Annotation Results via Ensemble Classifiers

tRNAScan‐SE is a tRNA detection program that is widely used for tRNA annotation; however, the false positive rate of tRNAScan‐SE is unacceptable for large sequences. Here, we used a machine learning method to try to improve the tRNAScan‐SE results. A new predictor, tRNA‐Predict, was designed. We obtained real and pseudo‐tRNA sequences as training data sets using tRNAScan‐SE and constructed three different tRNA feature sets. We then set up an ensemble classifier, LibMutil, to predict tRNAs from the training data. The positive data set of 623 tRNA sequences was obtained from tRNAdb 2009 and the negative data set was the false positive tRNAs predicted by tRNAscan‐SE. Our in silico experiments revealed a prediction accuracy rate of 95.1 % for tRNA‐Predict using 10‐fold cross‐validation. tRNA‐Predict was developed to distinguish functional tRNAs from pseudo‐tRNAs rather than to predict tRNAs from a genome‐wide scan. However, tRNA‐Predict can work with the output of tRNAscan‐SE, which is a genome‐wide scanning method, to improve the tRNAscan‐SE annotation results. The tRNA‐Predict web server is accessible at http://datamining.xmu.edu.cn/∼gjs/tRNA‐Predict.

Representation of fluctuation features in pathological knee joint vibroarthrographic signals using kernel density modeling method

This article applies advanced signal processing and computational methods to study the subtle fluctuations in knee joint vibroarthrographic (VAG) signals. Two new features are extracted to characterize the fluctuations of VAG signals. The fractal scaling index parameter is computed using the detrended fluctuation analysis algorithm to describe the fluctuations associated with intrinsic correlations in the VAG signal. The averaged envelope amplitude feature measures the difference between the upper and lower envelopes averaged over an entire VAG signal. Statistical analysis with the Kolmogorov-Smirnov test indicates that both of the fractal scaling index (p=0.0001) and averaged envelope amplitude (p=0.0001) features are significantly different between the normal and pathological signal groups. The bivariate Gaussian kernels are utilized for modeling the densities of normal and pathological signals in the two-dimensional feature space. Based on the feature densities estimated, the Bayesian decision rule makes better signal classifications than the least-squares support vector machine, with the overall classification accuracy of 88% and the area of 0.957 under the receiver operating characteristic (ROC) curve. Such VAG signal classification results are better than those reported in the state-of-the-art literature. The fluctuation features of VAG signals developed in the present study can provide useful information on the pathological conditions of degenerative knee joints. Classification results demonstrate the effectiveness of the kernel feature density modeling method for computer-aided VAG signal analysis.

Effective Dysphonia Detection Using Feature Dimension Reduction and Kernel Density Estimation for Patients with Parkinson's Disease

Detection of dysphonia is useful for monitoring the progression of phonatory impairment for patients with Parkinson’s disease (PD), and also helps assess the disease severity. This paper describes the statistical pattern analysis methods to study different vocal measurements of sustained phonations. The feature dimension reduction procedure was implemented by using the sequential forward selection (SFS) and kernel principal component analysis (KPCA) methods. Four selected vocal measures were projected by the KPCA onto the bivariate feature space, in which the class-conditional feature densities can be approximated with the nonparametric kernel density estimation technique. In the vocal pattern classification experiments, Fisher’s linear discriminant analysis (FLDA) was applied to perform the linear classification of voice records for healthy control subjects and PD patients, and the maximum a posteriori (MAP) decision rule and support vector machine (SVM) with radial basis function kernels were employed for the nonlinear classification tasks. Based on the KPCA-mapped feature densities, the MAP classifier successfully distinguished 91.8% voice records, with a sensitivity rate of 0.986, a specificity rate of 0.708, and an area value of 0.94 under the receiver operating characteristic (ROC) curve. The diagnostic performance provided by the MAP classifier was superior to those of the FLDA and SVM classifiers. In addition, the classification results indicated that gender is insensitive to dysphonia detection, and the sustained phonations of PD patients with minimal functional disability are more difficult to be correctly identified.

Removal of artifacts in knee joint vibroarthrographic signals using ensemble empirical mode decomposition and detrended fluctuation analysis.

High-resolution knee joint vibroarthrographic (VAG) signals can help physicians accurately evaluate the pathological condition of a degenerative knee joint, in order to prevent unnecessary exploratory surgery. Artifact cancellation is vital to preserve the quality of VAG signals prior to further computer-aided analysis. This paper describes a novel method that effectively utilizes ensemble empirical mode decomposition (EEMD) and detrended fluctuation analysis (DFA) algorithms for the removal of baseline wander and white noise in VAG signal processing. The EEMD method first successively decomposes the raw VAG signal into a set of intrinsic mode functions (IMFs) with fast and low oscillations, until the monotonic baseline wander remains in the last residue. Then, the DFA algorithm is applied to compute the fractal scaling index parameter for each IMF, in order to identify the anti-correlation and the long-range correlation components. Next, the DFA algorithm can be used to identify the anti-correlated and the long-range correlated IMFs, which assists in reconstructing the artifact-reduced VAG signals. Our experimental results showed that the combination of EEMD and DFA algorithms was able to provide averaged signal-to-noise ratio (SNR) values of 20.52 dB (standard deviation: 1.14 dB) and 20.87 dB (standard deviation: 1.89 dB) for 45 normal signals in healthy subjects and 20 pathological signals in symptomatic patients, respectively. The combination of EEMD and DFA algorithms can ameliorate the quality of VAG signals with great SNR improvements over the raw signal, and the results were also superior to those achieved by wavelet matching pursuit decomposition and time-delay neural filter.

Finding the Best Classification Threshold in Imbalanced Classification

Abstract Classification with imbalanced class distributions is a major problem in machine learning. Researchers have given considerable attention to the applications in many real-world scenarios. Although several works have utilized the area under the receiver operating characteristic (ROC) curve to select potentially optimal classifiers in imbalanced classifications, limited studies have been devoted to finding the classification threshold for testing or unknown datasets. In general, the classification threshold is simply set to 0.5, which is usually unsuitable for an imbalanced classification. In this study, we analyze the drawbacks of using ROC as the sole measure of imbalance in data classification problems. In addition, a novel framework for finding the best classification threshold is proposed. Experiments with SCOP v.1.53 data reveal that, with the default threshold set to 0.5, our proposed framework demonstrated a 20.63% improvement in terms of F-score compared with that of more commonly used methods. The findings suggest that the proposed framework is both effective and efficient. A web server and software tools are available via http://datamining.xmu.edu.cn/prht/ or http://prht.sinaapp.com/ .

Quantification of knee vibroarthrographic signal irregularity associated with patellofemoral joint cartilage pathology based on entropy and envelope amplitude measures

BACKGROUND AND OBJECTIVE Injury of knee joint cartilage may result in pathological vibrations between the articular surfaces during extension and flexion motions. The aim of this paper is to analyze and quantify vibroarthrographic (VAG) signal irregularity associated with articular cartilage degeneration and injury in the patellofemoral joint. METHODS The symbolic entropy (SyEn), approximate entropy (ApEn), fuzzy entropy (FuzzyEn), and the mean, standard deviation, and root-mean-squared (RMS) values of the envelope amplitude, were utilized to quantify the signal fluctuations associated with articular cartilage pathology of the patellofemoral joint. The quadratic discriminant analysis (QDA), generalized logistic regression analysis (GLRA), and support vector machine (SVM) methods were used to perform signal pattern classifications. RESULTS The experimental results showed that the patients with cartilage pathology (CP) possess larger SyEn and ApEn, but smaller FuzzyEn, over the statistical significance level of the Wilcoxon rank-sum test (p<0.01), than the healthy subjects (HS). The mean, standard deviation, and RMS values computed from the amplitude difference between the upper and lower signal envelopes are also consistently and significantly larger (p<0.01) for the group of CP patients than for the HS group. The SVM based on the entropy and envelope amplitude features can provide superior classification performance as compared with QDA and GLRA, with an overall accuracy of 0.8356, sensitivity of 0.9444, specificity of 0.8, Matthews correlation coefficient of 0.6599, and an area of 0.9212 under the receiver operating characteristic curve. CONCLUSIONS The SyEn, ApEn, and FuzzyEn features can provide useful information about pathological VAG signal irregularity based on different entropy metrics. The statistical parameters of signal envelope amplitude can be used to characterize the temporal fluctuations related to the cartilage pathology.

Adding irrelevant information to the content prime reduces the prime-induced unmasking effect on speech recognition.

Presenting the early part of a nonsense sentence in quiet improves recognition of the last keyword of the sentence in a masker, especially a speech masker. This priming effect depends on higher-order processing of the prime information during target-masker segregation. This study investigated whether introducing irrelevant content information into the prime reduces the priming effect. The results showed that presenting the first four syllables (not including the second and third keywords) of the three-keyword target sentence in quiet significantly improved recognition of the second and third keywords in a two-talker-speech masker but not a noise masker, relative to the no-priming condition. Increasing the prime content from four to eight syllables (including the first and second keywords of the target sentence) further improved recognition of the third keyword in either the noise or speech masker. However, if the last four syllables of the eight-syllable prime were replaced by four irrelevant syllables (which did not occur in the target sentence), all the prime-induced speech-recognition improvements disappeared. Thus, knowing the early part of the target sentence mainly reduces informational masking of target speech, possibly by helping listeners attend to the target speech. Increasing the informative content of the prime further improves target-speech recognition probably by reducing the processing load. The reduction of the priming effect by adding irrelevant information to the prime is not due to introducing additional masking of the target speech.

Analysis and classification of stride patterns associated with children development using gait signal dynamics parameters and ensemble learning algorithms

Measuring stride variability and dynamics in children is useful for the quantitative study of gait maturation and neuromotor development in childhood and adolescence. In this paper, we computed the sample entropy (SampEn) and average stride interval (ASI) parameters to quantify the stride series of 50 gender-matched children participants in three age groups. We also normalized the SampEn and ASI values by leg length and body mass for each participant, respectively. Results show that the original and normalized SampEn values consistently decrease over the significance level of the Mann-Whitney U test (p < 0.01) in children of 3–14 years old, which indicates the stride irregularity has been significantly ameliorated with the body growth. The original and normalized ASI values are also significantly changing when comparing between any two groups of young (aged 3–5 years), middle (aged 6–8 years), and elder (aged 10–14 years) children. Such results suggest that healthy children may better modulate their gait cadence rhythm with the development of their musculoskeletal and neurological systems. In addition, the AdaBoost.M2 and Bagging algorithms were used to effectively distinguish the childrens gait patterns. These ensemble learning algorithms both provided excellent gait classification results in terms of overall accuracy (≥90%), recall (≥0.8), and precision (≥0.8077).

Dysphonic Voice Pattern Analysis of Patients in Parkinson’s Disease Using Minimum Interclass Probability Risk Feature Selection and Bagging Ensemble Learning Methods

Analysis of quantified voice patterns is useful in the detection and assessment of dysphonia and related phonation disorders. In this paper, we first study the linear correlations between 22 voice parameters of fundamental frequency variability, amplitude variations, and nonlinear measures. The highly correlated vocal parameters are combined by using the linear discriminant analysis method. Based on the probability density functions estimated by the Parzen-window technique, we propose an interclass probability risk (ICPR) method to select the vocal parameters with small ICPR values as dominant features and compare with the modified Kullback-Leibler divergence (MKLD) feature selection approach. The experimental results show that the generalized logistic regression analysis (GLRA), support vector machine (SVM), and Bagging ensemble algorithm input with the ICPR features can provide better classification results than the same classifiers with the MKLD selected features. The SVM is much better at distinguishing normal vocal patterns with a specificity of 0.8542. Among the three classification methods, the Bagging ensemble algorithm with ICPR features can identify 90.77% vocal patterns, with the highest sensitivity of 0.9796 and largest area value of 0.9558 under the receiver operating characteristic curve. The classification results demonstrate the effectiveness of our feature selection and pattern analysis methods for dysphonic voice detection and measurement.

Measuring signal fluctuations in gait rhythm time series of patients with Parkinson's disease using entropy parameters

Abstract Gait rhythm disturbances due to abnormal strides indicate the degenerative mobility regulation of motor neurons affected by Parkinsons disease (PD). The aim of this work is to compute the approximate entropy (ApEn), normalized symbolic entropy (NSE), and signal turns count (STC) parameters for the measurements of stride fluctuations in PD. Generalized linear regression analysis (GLRA) and support vector machine (SVM) techniques were employed to implement nonlinear gait pattern classifications. The classification performance was evaluated in terms of overall accuracy, sensitivity, specificity, precision, Matthews correlation coefficient (MCC), and area under the receiver operating characteristic (ROC) curve. Our experimental results indicated that the ApEn, NSE, and STC parameters computed from the stride series of PD patients were all significantly larger (Wilcoxon rank-sum test: p