Classification of EMG signals using PSO optimized SVM for diagnosis of neuromuscular disorders

Design and fuzzy logic control of an active wrist orthosis

People who perform excessive wrist movements throughout the day because of their professions have a higher risk of developing lateral and medial epicondylitis. If proper precautions are not taken against these diseases, serious consequences such as job loss and early retirement can occur. In this study, the design and control of an active wrist orthosis that is mobile, powerful and lightweight is presented as a means to avoid the occurrence and/or for the treatment of repetitive strain injuries in an effective manner. The device has an electromyography-based control strategy so that the user’s intention always comes first. In fact, the device–user interaction is mainly activated by the electromyography signals measured from the forearm muscles that are responsible for the extension and flexion wrist movements. Contractions of the muscles are detected using surface electromyography sensors, and the desired quantity of the velocity value of the wrist is extracted from a fuzzy logic controller. Then, the actuator system of the device comes into play by conveying the necessary motion support to the wrist. Experimental studies show that the presented device actually reduces the demand on the muscles involved in repetitive strain injuries while performing challenging daily life activities including extension and flexion wrist motions.

Quality Assessment of Gentiana rigescens from Different Geographical Origins Using FT-IR Spectroscopy Combined with HPLC

Gentiana rigescens is a precious herbal medicine in China because of its liver-protective and choleretic effects. A method for the qualitative identification and quantitative evaluation of G. rigescens from Yunnan Province, China, has been developed employing Fourier transform infrared (FT-IR) spectroscopy and high performance liquid chromatography (HPLC) with the aid of chemometrics such as partial least squares discriminant analysis (PLS-DA) and support vector machines (SVM) regression. Our results indicated that PLS-DA model could efficiently discriminate G. rigescens from different geographical origins. It was found that the samples which could not be determined accurately were in the margin or outside of the 95% confidence ellipses. Moreover, the result implied that geographical origins variation of root samples were more obvious than that of stems and leaves. The quantitative analysis was based on gentiopicroside content which was the main active constituent in G. rigescens. For the prediction of gentiopicroside, the performances of model based on the parameters selected through grid search algorithm (GS) with seven-fold cross validation were better than those based on genetic algorithm (GA) and particle swarm optimization algorithm (PSO). For the SVM-GS model, the result was satisfactory. FT-IR spectroscopy coupled with PLS-DA and SVM-GS can be an alternative strategy for qualitative identification and quantitative evaluation of G. rigescens.

Real-time eutrophication status evaluation of coastal waters using support vector machine with grid search algorithm

The development of techniques for real-time monitoring of the eutrophication status of coastal waters is of great importance for realizing potential cost savings in coastal monitoring programs and providing timely advice for marine health management. In this study, a GS optimized SVM was proposed to model relationships between 6 easily measured parameters (DO, Chl-a, C1, C2, C3 and C4) and the TRIX index for rapidly assessing marine eutrophication states of coastal waters. The good predictive performance of the developed method was indicated by the R² between the measured and predicted values (0.92 for the training dataset and 0.91 for the validation dataset) at a 95% confidence level. The classification accuracy of the eutrophication status was 86.5% for the training dataset and 85.6% for the validation dataset. The results indicated that it is feasible to develop an SVM technique for timely evaluation of the eutrophication status by easily measured parameters.

Nanoparticle classification in wide-field interferometric microscopy by supervised learning from model

Interference-enhanced wide-field nanoparticle imaging is a highly sensitive technique that has found numerous applications in labeled and label-free subdiffraction-limited pathogen detection. It also provides unique opportunities for nanoparticle classification upon detection. More specifically, the nanoparticle defocus images result in a particle-specific response that can be of great utility for nanoparticle classification, particularly based on type and size. In this work, we combine a model-based supervised learning algorithm with a wide-field common-path interferometric microscopy method to achieve accurate nanoparticle classification. We verify our classification schemes experimentally by blindly detecting gold and polystyrene nanospheres, and then classifying them in terms of type and size.

Geographical traceability of wild Boletus edulis based on data fusion of FT-MIR and ICP-AES coupled with data mining methods (SVM)

In this work, the data fusion strategy of Fourier transform mid infrared (FT-MIR) spectroscopy and inductively coupled plasma-atomic emission spectrometry (ICP-AES) was used in combination with Support Vector Machine (SVM) to determine the geographic origin of Boletus edulis collected from nine regions of Yunnan Province in China. Firstly, competitive adaptive reweighted sampling (CARS) was used for selecting an optimal combination of key wavenumbers of second derivative FT-MIR spectra, and thirteen elements were sorted with variable importance in projection (VIP) scores. Secondly, thirteen subsets of multi-elements with the best VIP score were generated and each subset was used to fuse with FT-MIR. Finally, the classification models were established by SVM, and the combination of parameter C and γ (gamma) of SVM models was calculated by the approaches of grid search (GS) and genetic algorithm (GA). The results showed that both GS-SVM and GA-SVM models achieved good performances based on the #9 subset and the prediction accuracy in calibration and validation sets of the two models were 81.40% and 90.91%, correspondingly. In conclusion, it indicated that the data fusion strategy of FT-MIR and ICP-AES coupled with the algorithm of SVM can be used as a reliable tool for accurate identification of B. edulis, and it can provide a useful way of thinking for the quality control of edible mushrooms.

A Density-Based Clustering Approach to Motor Unit Potential Characterizations to Support Diagnosis of Neuromuscular Disorders

Electrophysiological muscle classification involves characterization of extracted motor unit potentials (MUPs) followed by the aggregation of these MUP characterizations. Existing techniques consider three classes (i.e., myopathic, neurogenic, and normal) for both MUP characterization and electrophysiological muscle classification. However, diseased-induced MUP changes are continuous in nature, which make it difficult to find distinct boundaries between normal, myopathic, and neurogenic MUPs. Hence, MUP characterization based on more than three classes is better able to represent the various effects of disease. Here, a novel, electrophysio- logical muscle classification system is proposed, which considers a dynamic number of classes for characterizing MUPs. To this end, a clustering algorithm called neighbor- hood distances entropy consistency is proposed to find clusters with arbitrary shapes and densities in an MUP feature space. These clusters represent several concepts of MUP normality and abnormality and are used for MUP characterization instead of the conventional three classes. An examined muscle is then classified by embedding its MUP characterizations in a feature vector fed to an ensemble of support vector machine and nearest neighbor classifiers. For 103 sets of MUPs recorded in tibialis anterior muscles, the proposed system had a 97% electro-physiological muscle classification accuracy, which is significantly higher than in previous works.

An EMG-based feature extraction method using a normalized weight vertical visibility algorithm for myopathy and neuropathy detection

Background

Electromyography (EMG) signals recorded from healthy, myopathic, and amyotrophic lateral sclerosis (ALS) subjects are nonlinear, non-stationary, and similar in the time domain and the frequency domain. Therefore, it is difficult to classify these various statuses.

Methods

This study proposes an EMG-based feature extraction method based on a normalized weight vertical visibility algorithm (NWVVA) for myopathy and ALS detection. In this method, sampling points or nodes based on sampling theory are extracted, and features are derived based on relations among the vertical visibility nodes with their amplitude differences as weights. The features are calculated via selective statistical mechanics and measurements, and the obtained features are assembled into a feature matrix as classifier input. Finally, powerful classifiers, such as k-nearest neighbor, multilayer perceptron neural network, and support vector machine classifiers, are utilized to differentiate signals of healthy, myopathy, and ALS cases.

Results

Performance evaluation experiments are carried out, and the results revealed 98.36% accuracy, which corresponds to approximately a 2% improvement compared with conventional methods.

Conclusions

An EMG-based feature extraction method using a NWVVA is proposed and implemented to detect healthy, ALS, and myopathy statuses.

Features extraction and multi-classification of sEMG using a GPU-Accelerated GA/MLP hybrid algorithm

BACKGROUND

Surface electromyography (sEMG) signal is the combined effect of superficial muscle EMG and neural electrical activity. In recent years, researchers did large amount of human-machine system studies by using the physiological signals as control signals.

OBJECTIVE

To develop and test a new multi-classification method to improve performance of analyzing sEMG signals based on public sEMG dataset.

METHODS

First, ten features were selected as candidate features. Second, a genetic algorithm (GA) was applied to select representative features from the initial ten candidates. Third, a multi-layer perceptron (MLP) classifier was trained by the selected optimal features. Last, the trained classifier was used to predict the classes of sEMG signals. A special graphics processing unit (GPU) was used to speed up the learning process.

RESULTS

Experimental results show that the classification accuracy of the new method reached higher than 90%. Comparing to other previously reported results, using the new method yielded higher performance.

CONCLUSIONS

The proposed features selection method is effective and the classification result is accurate. In addition, our method could have practical application value in medical prosthetics and the potential to improve robustness of myoelectric pattern recognition.

A two-dimensional matrix image based feature extraction method for classification of sEMG: A comparative analysis based on SVM, KNN and RBF-NN

BACKGROUND

The computer mouse is an important human-computer interaction device. But patients with physical finger disability are unable to operate this device. Surface EMG (sEMG) can be monitored by electrodes on the skin surface and is a reflection of the neuromuscular activities. Therefore, we can control limbs auxiliary equipment by utilizing sEMG classification in order to help the physically disabled patients to operate the mouse.

OBJECTIVE

To develop a new a method to extract sEMG generated by finger motion and apply novel features to classify sEMG.

METHODS

A window-based data acquisition method was presented to extract signal samples from sEMG electordes. Afterwards, a two-dimensional matrix image based feature extraction method, which differs from the classical methods based on time domain or frequency domain, was employed to transform signal samples to feature maps used for classification. In the experiments, sEMG data samples produced by the index and middle fingers at the click of a mouse button were separately acquired. Then, characteristics of the samples were analyzed to generate a feature map for each sample. Finally, the machine learning classification algorithms (SVM, KNN, RBF-NN) were employed to classify these feature maps on a GPU.

RESULTS

The study demonstrated that all classifiers can identify and classify sEMG samples effectively. In particular, the accuracy of the SVM classifier reached up to 100%.

CONCLUSIONS

The signal separation method is a convenient, efficient and quick method, which can effectively extract the sEMG samples produced by fingers. In addition, unlike the classical methods, the new method enables to extract features by enlarging sample signals' energy appropriately. The classical machine learning classifiers all performed well by using these features.

A Novel Hybrid Feature Selection Model for Classification of Neuromuscular Dystrophies Using Bhattacharyya Coefficient, Genetic Algorithm and Radial Basis Function Based Support Vector Machine

An accurate classification of neuromuscular disorders is important in providing proper treatment facilities to the patients. Recently, the microarray technology is employed to monitor the level of activity or expression of large number of genes simultaneously. The gene expression data derived from the microarray experiment usually involve a large number of genes but a very few number of samples. There is a need to reduce the dimension of gene expression data which intends to find a small set of discriminative genes that accurately classifies the samples of various kinds of diseases. So, our goal is to find a small subset of genes which ensures the accurate classification of neuromuscular disorders. In the present paper, we propose a novel hybrid feature selection model for classification of neuromuscular disorders. The process of feature selection is done in two phases by integrating Bhattacharyya coefficient and genetic algorithm (GA). In the first phase, we find Bhattacharyya coefficient to choose a candidate gene subset by removing the most redundant genes. In the second phase, the target gene subset is created by selecting the most discriminative gene subset by applying GA wherein the fitness function is calculated using radial basis function support vector machine (RBF SVM). The proposed hybrid algorithm is applied on two publicly available microarray neuromuscular disorders datasets. The results are compared with two individual techniques of feature selection, namely Bhattacharyya coefficient and GA, and one integrated technique, i.e., Bhattacharyya-GA wherein the fitness function of GA is calculated using four other classifiers, which shows that the proposed integrated method is capable of giving the better classification accuracy.

A Review of Classification Techniques of EMG Signals during Isotonic and Isometric Contractions

In recent years, there has been major interest in the exposure to physical therapy during rehabilitation. Several publications have demonstrated its usefulness in clinical/medical and human machine interface (HMI) applications. An automated system will guide the user to perform the training during rehabilitation independently. Advances in engineering have extended electromyography (EMG) beyond the traditional diagnostic applications to also include applications in diverse areas such as movement analysis. This paper gives an overview of the numerous methods available to recognize motion patterns of EMG signals for both isotonic and isometric contractions. Various signal analysis methods are compared by illustrating their applicability in real-time settings. This paper will be of interest to researchers who would like to select the most appropriate methodology in classifying motion patterns, especially during different types of contractions. For feature extraction, the probability density function (PDF) of EMG signals will be the main interest of this study. Following that, a brief explanation of the different methods for pre-processing, feature extraction and classifying EMG signals will be compared in terms of their performance. The crux of this paper is to review the most recent developments and research studies related to the issues mentioned above.

Real-time seizure prediction using RLS filtering and interpolated histogram feature based on hybrid optimization algorithm of Bayesian classifier and Hunting search

BACKGROUND AND OBJECTIVES

Epileptic seizure prediction using EEG signal analysis is an important application for drug therapy and pediatric patient monitoring. Time series estimation to obtain the future samples of EEG signal has vital role for detecting seizure attack. In this paper, a novel density-based real-time seizure prediction algorithm based on a trained offline seizure detection algorithm is proposed.

METHODS

In the offline seizure detection procedure, after signal preprocessing, histogram-based statistical features are extracted from signal probability distribution. By defining a deterministic polynomial model on the normalized histogram, a novel syntactic feature that is named Interpolated Histogram Feature (IHF) is proposed. Moreover, with this feature, Seizure Distribution Model (SDM) as a descriptor of the seizure and non-seizure signals is presented. By using a novel hybrid optimization algorithm based on Bayesian classifier and Hunting Search (HuS) algorithm, the optimal features are selected. To detect the seizure attacks in the online mode, a Multi-Layer Perceptron (MLP) classifier is trained with the optimal features in the offline procedure. For online prediction, the enhanced Recursive Least Square (RLS) filter is applied to estimate sample-by-sample of the EEG signal. Also, a density-based signal tracking scenario is introduced to update and tune the parameters of RLS filtering algorithm.

RESULTS

Our prediction algorithm is evaluated on 104 hours of EEG signals recorded from 23 pediatric patients. Our online signal prediction algorithm provides the accuracy rate of 86.56% and precision rate of 86.53% simultaneously using the trained MLP classifier from the offline mode. The recall rate of seizure prediction is 97.27% and the false prediction rate of 0.00215 per hour is achieved as well. Ultimately, the future samples of EEG signal are estimated, and the time of seizure signal prediction is also converged to 6.64 seconds.

CONCLUSION

In our proposed real-time algorithm, by implementing a density-based signal tracking scenario, the future samples of signal with suitable time is predicted and the seizure is detected based on the optimal features from the IHF and histogram-based statistical features with acceptable performance.

Congestive heart failure detection using random forest classifier

BACKGROUND AND OBJECTIVES

Automatic electrocardiogram (ECG) heartbeat classification is substantial for diagnosing heart failure. The aim of this paper is to evaluate the effect of machine learning methods in creating the model which classifies normal and congestive heart failure (CHF) on the long-term ECG time series.

METHODS

The study was performed in two phases: feature extraction and classification phase. In feature extraction phase, autoregressive (AR) Burg method is applied for extracting features. In classification phase, five different classifiers are examined namely, C4.5 decision tree, k-nearest neighbor, support vector machine, artificial neural networks and random forest classifier. The ECG signals were acquired from BIDMC Congestive Heart Failure and PTB Diagnostic ECG databases and classified by applying various experiments.

RESULTS

The experimental results are evaluated in several statistical measures (sensitivity, specificity, accuracy, F-measure and ROC curve) and showed that the random forest method gives 100% classification accuracy.

CONCLUSIONS

Impressive performance of random forest method proves that it plays significant role in detecting congestive heart failure (CHF) and can be valuable in expressing knowledge useful in medicine.

Basic Hand Gestures Classification Based on Surface Electromyography

This paper presents an innovative classification system for hand gestures using 2-channel surface electromyography analysis. The system developed uses the Support Vector Machine classifier, for which the kernel function and parameter optimisation are conducted additionally by the Cuckoo Search swarm algorithm. The system developed is compared with standard Support Vector Machine classifiers with various kernel functions. The average classification rate of 98.12% has been achieved for the proposed method.

Dissolved oxygen content prediction in crab culture using a hybrid intelligent method

A precise predictive model is needed to obtain a clear understanding of the changing dissolved oxygen content in outdoor crab ponds, to assess how to reduce risk and to optimize water quality management. The uncertainties in the data from multiple sensors are a significant factor when building a dissolved oxygen content prediction model. To increase prediction accuracy, a new hybrid dissolved oxygen content forecasting model based on the radial basis function neural networks (RBFNN) data fusion method and a least squares support vector machine (LSSVM) with an optimal improved particle swarm optimization(IPSO) is developed. In the modelling process, the RBFNN data fusion method is used to improve information accuracy and provide more trustworthy training samples for the IPSO-LSSVM prediction model. The LSSVM is a powerful tool for achieving nonlinear dissolved oxygen content forecasting. In addition, an improved particle swarm optimization algorithm is developed to determine the optimal parameters for the LSSVM with high accuracy and generalizability. In this study, the comparison of the prediction results of different traditional models validates the effectiveness and accuracy of the proposed hybrid RBFNN-IPSO-LSSVM model for dissolved oxygen content prediction in outdoor crab ponds.

Classification of Motor Imagery EEG Signals with Support Vector Machines and Particle Swarm Optimization

Support vector machines are powerful tools used to solve the small sample and nonlinear classification problems, but their ultimate classification performance depends heavily upon the selection of appropriate kernel and penalty parameters. In this study, we propose using a particle swarm optimization algorithm to optimize the selection of both the kernel and penalty parameters in order to improve the classification performance of support vector machines. The performance of the optimized classifier was evaluated with motor imagery EEG signals in terms of both classification and prediction. Results show that the optimized classifier can significantly improve the classification accuracy of motor imagery EEG signals.

Medical Decision Support System for Diagnosis of Heart Arrhythmia using DWT and Random Forests Classifier

In this study, Random Forests (RF) classifier is proposed for ECG heartbeat signal classification in diagnosis of heart arrhythmia. Discrete wavelet transform (DWT) is used to decompose ECG signals into different successive frequency bands. A set of different statistical features were extracted from the obtained frequency bands to denote the distribution of wavelet coefficients. This study shows that RF classifier achieves superior performances compared to other decision tree methods using 10-fold cross-validation for the ECG datasets and the obtained results suggest that further significant improvements in terms of classification accuracy can be accomplished by the proposed classification system. Accurate ECG signal classification is the major requirement for detection of all arrhythmia types. Performances of the proposed system have been evaluated on two different databases, namely MIT-BIH database and St. -Petersburg Institute of Cardiological Technics 12-lead Arrhythmia Database. For MIT-BIH database, RF classifier yielded an overall accuracy 99.33 % against 98.44 and 98.67 % for the C4.5 and CART classifiers, respectively. For St. -Petersburg Institute of Cardiological Technics 12-lead Arrhythmia Database, RF classifier yielded an overall accuracy 99.95 % against 99.80 % for both C4.5 and CART classifiers, respectively. The combined model with multiscale principal component analysis (MSPCA) de-noising, discrete wavelet transform (DWT) and RF classifier also achieves better performance with the area under the receiver operating characteristic (ROC) curve (AUC) and F-measure equal to 0.999 and 0.993 for MIT-BIH database and 1 and 0.999 for and St. -Petersburg Institute of Cardiological Technics 12-lead Arrhythmia Database, respectively. Obtained results demonstrate that the proposed system has capacity for reliable classification of ECG signals, and to assist the clinicians for making an accurate diagnosis of cardiovascular disorders (CVDs).

A novel approach for SEMG signal classification with adaptive local binary patterns

Feature extraction plays a major role in the pattern recognition process, and this paper presents a novel feature extraction approach, adaptive local binary pattern (aLBP). aLBP is built on the local binary pattern (LBP), which is an image processing method, and one-dimensional local binary pattern (1D-LBP). In LBP, each pixel is compared with its neighbors. Similarly, in 1D-LBP, each data in the raw is judged against its neighbors. 1D-LBP extracts feature based on local changes in the signal. Therefore, it has high a potential to be employed in medical purposes. Since, each action or abnormality, which is recorded in SEMG signals, has its own pattern, and via the 1D-LBP these (hidden) patterns may be detected. But, the positions of the neighbors in 1D-LBP are constant depending on the position of the data in the raw. Also, both LBP and 1D-LBP are very sensitive to noise. Therefore, its capacity in detecting hidden patterns is limited. To overcome these drawbacks, aLBP was proposed. In aLBP, the positions of the neighbors and their values can be assigned adaptively via the down-sampling and the smoothing coefficients. Therefore, the potential to detect (hidden) patterns, which may express an illness or an action, is really increased. To validate the proposed feature extraction approach, two different datasets were employed. Achieved accuracies by the proposed approach were higher than obtained results by employed popular feature extraction approaches and the reported results in the literature. Obtained accuracy results were brought out that the proposed method can be employed to investigate SEMG signals. In summary, this work attempts to develop an adaptive feature extraction scheme that can be utilized for extracting features from local changes in different categories of time-varying signals.

Alterations in lower limb multimuscle activation patterns during stair climbing in female total knee arthroplasty patients

Total knee arthroplasty (TKA) patients commonly experience neuromuscular adaptations that may affect stair climbing competence. This study identified multimuscle pattern (MMP) changes in postoperative female TKA patients during stair climbing with a support vector machine (SVM). It was hypothesized that TKA patients adopt temporal and spectral muscle activation characteristics indicative of muscle atrophy and cocontraction strategies. Nineteen female subjects [10 unilateral sex-specific TKAs, 62.2 ± 8.6 yr, body mass index (BMI) 28.2 ± 5.4 kg/m(2); 9 healthy control subjects, 61.4 ± 7.4 yr, BMI 25.6 ± 2.4 kg/m(2)] were recruited. Surface electromyograms (EMGs) were obtained for seven lower limb muscles of the affected limb of TKA subjects and a randomly assigned limb for control subjects during stair climbing. Stance phase (±30%) EMG data were wavelet transformed and normalized to total power. Data across all muscles were combined to form MMPs and analyzed with a SVM. Statistical analysis was performed with binomial tests, independent group t-tests, or independent group Mann-Whitney U-tests in SPSS (P < 0.05). SVM results indicated significantly altered muscle activation patterns in the TKA group for biceps femoris (recognition rate 84.2%), semitendinosus (recognition rate 73.7%), gastrocnemius (recognition rate 68.4%), and tibialis anterior (recognition rate 68.4%). Further analysis identified no significant differences in spectral activation characteristics between groups. Temporal adaptations, indicative of cocontraction strategies, were, however, evident in TKA MMPs. This approach may provide a valuable tool for clinical neuromuscular function assessment and rehabilitation monitoring.

Single-Channel EMG Classification With Ensemble-Empirical-Mode-Decomposition-Based ICA for Diagnosing Neuromuscular Disorders

An accurate and computationally efficient quantitative analysis of electromyography (EMG) signals plays an inevitable role in the diagnosis of neuromuscular disorders, prosthesis, and several related applications. Since it is often the case that the measured signals are the mixtures of electric potentials that emanate from surrounding muscles (sources), many EMG signal processing approaches rely on linear source separation techniques such as the independent component analysis (ICA). Nevertheless, naive implementations of ICA algorithms do not comply with the task of extracting the underlying sources from a single-channel EMG measurement. In this respect, the present work focuses on a classification method for neuromuscular disorders that deals with the data recorded using a single-channel EMG sensor. The ensemble empirical mode decomposition algorithm decomposes the single-channel EMG signal into a set of noise-canceled intrinsic mode functions, which in turn are separated by the FastICA algorithm. A reduced set of five time domain features extracted from the separated components are classified using the linear discriminant analysis, and the classification results are fine-tuned with a majority voting scheme. The performance of the proposed method has been validated with a clinical EMG database, which reports a higher classification accuracy (98%). The outcome of this study encourages possible extension of this approach to real settings to assist the clinicians in making correct diagnosis of neuromuscular disorders.

A Novel Feature Extraction Approach Using Window Function Capturing and QPSO-SVM for Enhancing Electronic Nose Performance

In this paper, a novel feature extraction approach which can be referred to as moving window function capturing (MWFC) has been proposed to analyze signals of an electronic nose (E-nose) used for detecting types of infectious pathogens in rat wounds. Meanwhile, a quantum-behaved particle swarm optimization (QPSO) algorithm is implemented in conjunction with support vector machine (SVM) for realizing a synchronization optimization of the sensor array and SVM model parameters. The results prove the efficacy of the proposed method for E-nose feature extraction, which can lead to a higher classification accuracy rate compared to other established techniques. Meanwhile it is interesting to note that different classification results can be obtained by changing the types, widths or positions of windows. By selecting the optimum window function for the sensor response, the performance of an E-nose can be enhanced.

A Hybrid Swarm Algorithm for optimizing glaucoma diagnosis

Glaucoma is among the most common causes of permanent blindness in human. Because the initial symptoms are not evident, mass screening would assist early diagnosis in the vast population. Such mass screening requires an automated diagnosis technique. Our proposed automation consists of pre-processing, optimal wavelet transformation, feature extraction, and classification modules. The hyper analytic wavelet transformation (HWT) based statistical features are extracted from fundus images. Because HWT preserves phase information, it is appropriate for feature extraction. The features are then classified by a Support Vector Machine (SVM) with a radial basis function (RBF) kernel. The filter coefficients of the wavelet transformation process and the SVM-RB width parameter are simultaneously tailored to best-fit the diagnosis by the hybrid Particle Swarm algorithm. To overcome premature convergence, a Group Search Optimizer (GSO) random searching (ranging) and area scanning behavior (around the optima) are embedded within the Particle Swarm Optimization (PSO) framework. We also embed a novel potential-area scanning as a preventive mechanism against premature convergence, rather than diagnosis and cure. This embedding does not compromise the generality and utility of PSO. In two 10-fold cross-validated test runs, the diagnostic accuracy of the proposed hybrid PSO exceeded that of conventional PSO. Furthermore, the hybrid PSO maintained the ability to explore even at later iterations, ensuring maturity in fitness.

Automated colon cancer detection using hybrid of novel geometric features and some traditional features

Automatic classification of colon into normal and malignant classes is complex due to numerous factors including similar colors in different biological constituents of histopathological imagery. Therefore, such techniques, which exploit the textural and geometric properties of constituents of colon tissues, are desired. In this paper, a novel feature extraction strategy that mathematically models the geometric characteristics of constituents of colon tissues is proposed. In this study, we also show that the hybrid feature space encompassing diverse knowledge about the tissues׳ characteristics is quite promising for classification of colon biopsy images. This paper thus presents a hybrid feature space based colon classification (HFS-CC) technique, which utilizes hybrid features for differentiating normal and malignant colon samples. The hybrid feature space is formed to provide the classifier different types of discriminative features such as features having rich information about geometric structure and image texture. Along with the proposed geometric features, a few conventional features such as morphological, texture, scale invariant feature transform (SIFT), and elliptic Fourier descriptors (EFDs) are also used to develop a hybrid feature set. The SIFT features are reduced using minimum redundancy and maximum relevancy (mRMR). Various kernels of support vector machines (SVM) are employed as classifiers, and their performance is analyzed on 174 colon biopsy images. The proposed geometric features have achieved an accuracy of 92.62%, thereby showing their effectiveness. Moreover, the proposed HFS-CC technique achieves 98.07% testing and 99.18% training accuracy. The better performance of HFS-CC is largely due to the discerning ability of the proposed geometric features and the developed hybrid feature space.

GECC: Gene Expression Based Ensemble Classification of Colon Samples

Gene expression deviates from its normal composition in case a patient has cancer. This variation can be used as an effective tool to find cancer. In this study, we propose a novel gene expressions based colon classification scheme (GECC) that exploits the variations in gene expressions for classifying colon gene samples into normal and malignant classes. Novelty of GECC is in two complementary ways. First, to cater overwhelmingly larger size of gene based data sets, various feature extraction strategies, like, chi-square, F-Score, principal component analysis (PCA) and minimum redundancy and maximum relevancy (mRMR) have been employed, which select discriminative genes amongst a set of genes. Second, a majority voting based ensemble of support vector machine (SVM) has been proposed to classify the given gene based samples. Previously, individual SVM models have been used for colon classification, however, their performance is limited. In this research study, we propose an SVM-ensemble based new approach for gene based classification of colon, wherein the individual SVM models are constructed through the learning of different SVM kernels, like, linear, polynomial, radial basis function (RBF), and sigmoid. The predicted results of individual models are combined through majority voting. In this way, the combined decision space becomes more discriminative. The proposed technique has been tested on four colon, and several other binary-class gene expression data sets, and improved performance has been achieved compared to previously reported gene based colon cancer detection techniques. The computational time required for the training and testing of 208 × 5,851 data set has been 591.01 and 0.019 s, respectively.

Distributed object detection with linear SVMs

In vision and learning, low computational complexity and high generalization are two important goals for video object detection. Low computational complexity here means not only fast speed but also less energy consumption. The sliding window object detection method with linear support vector machines (SVMs) is a general object detection framework. The computational cost is herein mainly paid in complex feature extraction and innerproduct-based classification. This paper first develops a distributed object detection framework (DOD) by making the best use of spatial-temporal correlation, where the process of feature extraction and classification is distributed in the current frame and several previous frames. In each framework, only subfeature vectors are extracted and the response of partial linear classifier (i.e., subdecision value) is computed. To reduce the dimension of traditional block-based histograms of oriented gradients (BHOG) feature vector, this paper proposes a cell-based HOG (CHOG) algorithm, where the features in one cell are not shared with overlapping blocks. Using CHOG as feature descriptor, we develop CHOG-DOD as an instance of DOD framework. Experimental results on detection of hand, face, and pedestrian in video show the superiority of the proposed method.