Diagnosis of Parkinson’s disease (PD) based on speech data has been proved to be an effective way in recent years. However, current researches just care about the feature extraction and classifier design, and do not consider the instance selection. Former research by authors showed that the instance selection can lead to improvement on classification accuracy. However, no attention is paid on the relationship between speech sample and feature until now. Therefore, a new diagnosis algorithm of PD is proposed in this paper by simultaneously selecting speech sample and feature based on relevant feature weighting algorithm and multiple kernel method, so as to find their synergy effects, thereby improving classification accuracy. Experimental results showed that this proposed algorithm obtained apparent improvement on classification accuracy. It can obtain mean classification accuracy of 82.5%, which was 30.5% higher than the relevant algorithm. Besides, the proposed algorithm detected the synergy effects of speech sample and feature, which is valuable for speech marker extraction.
How to extract high discriminative features that help classification from complex resting-state fMRI (rs-fMRI) data is the key to improving the accuracy of brain disease recognition such as schizophrenia. In this work, we use a weighted sparse model for brain network construction, and utilize the Kendall correlation coefficient (KCC) to extract the discriminative connectivity features for schizophrenia classification, which is conducted with the linear support vector machine. Experimental results based on the rs-fMRI of 57 schizophrenia patients and 64 healthy controls show that our proposed method is more effective (i.e., achieving a significantly higher classification accuracy, 81.82%) than other competing methods. Specifically, compared with the traditional network construction methods (Pearson’s correlation and sparse representation) and the commonly used feature selection methods (two-sample t-test and Least absolute shrinkage and selection operator (Lasso)), the algorithm proposed in this paper can more effectively extract the discriminative connectivity features between the schizophrenia patients and the healthy controls, and further improve the classification accuracy. At the same time, the discriminative connectivity features extracted in the work could be used as the potential clinical biomarkers to assist the identification of schizophrenia.
It is of great clinical significance in the differential diagnosis of primary central nervous system lymphoma (PCNSL) and glioblastoma (GBM) because there are enormous differences between them in terms of therapeutic regimens. In this paper, we propose a system based on sparse representation for automatic classification of PCNSL and GBM. The proposed system distinguishes the two tumors by using of the different texture detail information of the two tumors on T1 contrast magnetic resonance imaging (MRI) images. First, inspired by the process of radiomics, we designed a dictionary learning and sparse representation-based method to extract texture information, and with this approach, the tumors with different volume and shape were transformed into 968 quantitative texture features. Next, aiming at the problem of the redundancy in the extracted features, feature selection based on iterative sparse representation was set up to select some key texture features with high stability and discrimination. Finally, the selected key features are used for differentiation based on sparse representation classification (SRC) method. By using ten-fold cross-validation method, the differentiation based on the proposed approach presents accuracy of 96.36%, sensitivity 96.30%, and specificity 96.43%. Experimental results show that our approach not only effectively distinguish the two tumors but also has strong robustness in practical application since it avoids the process of parameter extraction on advanced MRI images.
The feature extraction and feature selection are the important issues in pattern recognition. Based on the geometric algebra representation of vector, a new feature extraction method using blade coefficient of geometric algebra was proposed in this study. At the same time, an improved differential evolution (DE) feature selection method was proposed to solve the elevated high dimension issue. The simple linear discriminant analysis was used as the classifier. The result of the 10-fold cross-validation (10 CV) classification of public breast cancer biomedical dataset was more than 96% and proved superior to that of the original features and traditional feature extraction method.
The purpose of our study is to evaluate the diagnostic performance of radiomics in multi-class discrimination of lymphadenopathy based on elastography and B-mode dual-modal ultrasound images. We retrospectively analyzed a total of 251 lymph nodes (89 benign lymph nodes, 70 lymphoma and 92 metastatic lymph nodes) from 248 patients, which were examined by both elastography and B-mode sonography. Firstly, radiomic features were extracted from multimodal ultrasound images, including shape features, intensity statistics features and gray-level co-occurrence matrix texture features. Secondly, three feature selection methods based on information theory were used on the radiomic features to select different subsets of radiomic features, consisting of conditional infomax feature extraction, conditional mutual information maximization, and double input symmetric relevance. Thirdly, the support vector machine classifier was performed for diagnosis of lymphadenopathy on each radiomic subsets. Finally, we fused the results from different modalities and different radiomic feature subsets with Adaboost to improve the performance of lymph node classification. The results showed that the accuracy and overall F1 score with five-fold cross-validation were 76.09%±1.41% and 75.88%±4.32%, respectively. Moreover, when considering on benign lymph nodes, lymphoma or metastatic lymph nodes respectively, the area under the receiver operating characteristic curve of multi-class classification were 0.77, 0.93 and 0.84, respectively. This study indicates that radiomic features derived from multimodal ultrasound images are benefit for diagnosis of lymphadenopathy. It is expected to be useful in clinical differentiation of lymph node diseases.
Signal classification is a key of brain-computer interface (BCI). In this paper, we present a new method for classifying the electroencephalogram (EEG) signals of which the features are heterogeneous. This method is called wrapped elastic net feature selection and classification. Firstly, we used the joint application of time-domain statistic, power spectral density (PSD), common spatial pattern (CSP) and autoregressive (AR) model to extract high-dimensional fused features of the preprocessed EEG signals. Then we used the wrapped method for feature selection. We fitted the logistic regression model penalized with elastic net on the training data, and obtained the parameter estimation by coordinate descent method. Then we selected best feature subset by using 10-fold cross-validation. Finally, we classified the test sample using the trained model. Data used in the experiment were the EEG data from international BCI Competition Ⅳ. The results showed that the method proposed was suitable for fused feature selection with high-dimension. For identifying EEG signals, it is more effective and faster, and can single out a more relevant subset to obtain a relatively simple model. The average test accuracy reached 81.78%.
The study was intended to introduce a novel method for aided diagnosis of cardiovascular diseases based on photoplethysmography (PPG). For this purpose, 40 volunteers were recruited in this study, of whom the physiological and pathological information was collected, including blood pressure and simultaneous PPG data on fingertips, by using a sphygmomanometer and a smart fingertip sensor. According to the PPG signal and its first and second derivatives, 52 features were defined and acquired. The Relief feature selection algorithm was performed to calculate the contribution of each feature to cardiovascular diseases. And then 10 core features which had the greatest contribution were selected as an optimal feature subset. Finally, the efficiency of the Relief feature selection algorithm was demonstrated by the results of k-nearest neighbor (kNN) and support vector machine (SVM) classifier applications of the features. The prediction accuracy of kNN model and SVM reached 66.67% and 83.33% respectively, indicating that: ① Age was the foremost feature for aided diagnosis of cardiovascular diseases; ② The optimal feature subset provided an important evaluation of cardiovascular health status. The obtained results showed that the application of the Relief feature selection algorithm provided high accuracy in aided diagnosis of cardiovascular diseases.
The early diagnosis of children with autism spectrum disorders (ASD) is essential. Electroencephalography (EEG) is one of most commonly used neuroimaging techniques as the most accessible and informative method. In this study, approximate entropy (ApEn), sample entropy (SaEn), permutation entropy (PeEn) and wavelet entropy (WaEn) were extracted from EEGs of ASD child and a control group, and Student's t-test was used to analyze between-group differences. Support vector machine (SVM) algorithm was utilized to build classification models for each entropy measure derived from different regions. Permutation test was applied in search for optimize subset of features, with which the SVM model achieved best performance. The results showed that the complexity of EEGs in children with autism was lower than that of the normal control group. Among all four entropies, WaEn got a better classification performance than others. Classification results vary in different regions, and the frontal lobe showed the best performance. After feature selection, six features were filtered out and the accuracy rate was increased to 84.55%, which can be convincing for assisting early diagnosis of autism.
Objective To propose a heart sound segmentation method based on multi-feature fusion network. Methods Data were obtained from the CinC/PhysioNet 2016 Challenge dataset (a total of 3 153 recordings from 764 patients, about 91.93% of whom were male, with an average age of 30.36 years). Firstly the features were extracted in time domain and time-frequency domain respectively, and reduced redundant features by feature dimensionality reduction. Then, we selected optimal features separately from the two feature spaces that performed best through feature selection. Next, the multi-feature fusion was completed through multi-scale dilated convolution, cooperative fusion, and channel attention mechanism. Finally, the fused features were fed into a bidirectional gated recurrent unit (BiGRU) network to heart sound segmentation results. Results The proposed method achieved precision, recall and F1 score of 96.70%, 96.99%, and 96.84% respectively. Conclusion The multi-feature fusion network proposed in this study has better heart sound segmentation performance, which can provide high-accuracy heart sound segmentation technology support for the design of automatic analysis of heart diseases based on heart sounds.
In order to improve the accuracy of classification with small amount of motor imagery training data on the development of brain-computer interface (BCI) systems, we proposed an analyzing method to automatically select the characteristic parameters based on correlation coefficient analysis. Throughout the five sample data of dataset Ⅳa from 2005 BCI Competition, we utilized short-time Fourier transform (STFT) and correlation coefficient calculation to reduce the number of primitive electroencephalogram dimension, then introduced feature extraction based on common spatial pattern (CSP) and classified by linear discriminant analysis (LDA). Simulation results showed that the average rate of classification accuracy could be improved by using correlation coefficient feature selection method than those without using this algorithm. Comparing with support vector machine (SVM) optimization features algorithm, the correlation coefficient analysis can lead better selection parameters to improve the accuracy of classification.