An efficient Computer Aided Decision Support System for breast cancer diagnosis using Echo State Network classifier

With the 3D printing rapidly expanding into various fields, 3D printers, as the equipment, should adopt a low-cost and small-sample fault diagnosis methods. A fault diagnosis method based on echo state networks (ESN) for 3D printers is proposed in this paper. A low-cost attitude sensor installed on the 3D printer is employed to collect raw fault data. Subsequently, feature extraction is carried out on the raw fault data. Using these features, ESN, as a shallow learning network, is modeled to diagnose faults of 3D printers. Experimental results show that the fault diagnosis method based on ESN still effective for 3D printers in low-cost and small-sample, which can make the fault recognition accuracy of 3D printer reach to 97.26%. Furthermore, contrast results indicated that the fault diagnosis accuracy of ESN is higher and most stable when compare with support vector machine (SVM), locality preserving projection support vector machine (LPPSVM) and principal component analysis support vector machine (PCASVM).

The paper applies recently developed smart approach for feature extraction from multi-dimensional data sets using Echo state networks (ESN) to the focalized spectra obtained from the acoustic camera multi-sensor measurements. The aim of the study is development of distance diagnostic system for prediction of wearing out of bearings. The procedure for initial features selection and features extraction from the focalized spectra was developed. Then the k-means clustering algorithm and Support vector machine (SVM) classifiers were applied to differentiate the tested bearings into two classes with respect to their condition (“Good” or “Bad”). The results using different dimensions of the extracted features space were compared.

In computer-aided detection of microcalcifications (MCs), the detection accuracy is often compromised by frequent occurrence of false positives (FPs), which can be attributed to a number of factors, including imaging noise, inhomogeneity in tissue background, linear structures, and artifacts in mammograms. In this study, the authors investigated a unified classification approach for combating the adverse effects of these heterogeneous factors for accurate MC detection. To accommodate FPs caused by different factors in a mammogram image, the authors developed a classification model to which the input features were adapted according to the image context at a detection location. For this purpose, the input features were defined in two groups, of which one group was derived from the image intensity pattern in a local neighborhood of a detection location, and the other group was used to characterize how a MC is different from its structural background. Owing to the distinctive effect of linear structures in the detector response, the authors introduced a dummy variable into the unified classifier model, which allowed the input features to be adapted according to the image context at a detection location (i.e., presence or absence of linear structures). To suppress the effect of inhomogeneity in tissue background, the input features were extracted from different domains aimed for enhancing MCs in a mammogram image. To demonstrate the flexibility of the proposed approach, the authors implemented the unified classifier model by two widely used machine learning algorithms, namely, a support vector machine (SVM) classifier and an Adaboost classifier. In the experiment, the proposed approach was tested for two representative MC detectors in the literature [difference-of-Gaussians (DoG) detector and SVM detector]. The detection performance was assessed using free-response receiver operating characteristic (FROC) analysis on a set of 141 screen-film mammogram (SFM) images (66 cases) and a set of 188 full-field digital mammogram (FFDM) images (95 cases). The FROC analysis results show that the proposed unified classification approach can significantly improve the detection accuracy of two MC detectors on both SFM and FFDM images. Despite the difference in performance between the two detectors, the unified classifiers can reduce their FP rate to a similar level in the output of the two detectors. In particular, with true-positive rate at 85%, the FP rate on SFM images for the DoG detector was reduced from 1.16 to 0.33 clusters/image (unified SVM) and 0.36 clusters/image (unified Adaboost), respectively; similarly, for the SVM detector, the FP rate was reduced from 0.45 clusters/image to 0.30 clusters/image (unified SVM) and 0.25 clusters/image (unified Adaboost), respectively. Similar FP reduction results were also achieved on FFDM images for the two MC detectors. The proposed unified classification approach can be effective for discriminating MCs from FPs caused by different factors (such as MC-like noise patterns and linear structures) in MC detection. The framework is general and can be applicable for further improving the detection accuracy of existing MC detectors.

Echo state networks (ESN), a type of reservoir computing (RC) architecture, are efficient and accurate artificial neural systems for time series processing and learning. An ESN consists of a core of recurrent neural networks, called a reservoir, with a small number of tunable parameters to generate a high-dimensional representation of an input, and a readout layer which is easily trained using regression to produce a desired output from the reservoir states. Certain computational tasks involve real-time calculation of high-order time correlations, which requires nonlinear transformation either in the reservoir or the readout layer. Traditional ESN employs a reservoir with sigmoid or tanh function neurons. In contrast, some types of biological neurons obey response curves that can be described as a product unit rather than a sum and threshold. Inspired by this class of neurons, we introduce a RC architecture with a reservoir of product nodes for time series computation. We find that the product RC shows many properties of standard ESN such as short-term memory and nonlinear capacity. On standard benchmarks for chaotic prediction tasks, the product RC maintains the performance of a standard nonlinear ESN while being more amenable to mathematical analysis. Our study provides evidence that such networks are powerful in highly nonlinear tasks owing to high-order statistics generated by the recurrent product node reservoir.

In all customer based industries, customer churn is considered as one of the most important and challenging concerns since it can lead to a serious profit loss. Therefore, developing accurate churn prediction models can significantly help Customer Relationship Management in planning effective retention campaigns and consequently helps in maximizing the profit of the service provider. In this paper, we propose the use of an Echo State Network (ESN) with a Support Vector Machine (SVM) training algorithm for predicting customer churn in telecommunication companies. The proposed approach is trained and tested based on two datasets: the first is a popular online available dataset while the second is obtained from a local service provider. Experiment results show that ESN with SVM readout outperform other popular machine learning models used in the literature for the same customer churn prediction problems.

SummaryIn order to improve the prediction accuracy of network traffic, a novel prediction model based on ensemble empirical mode decomposition and multiple models is proposed. In this study, ensemble empirical mode decomposition algorithm is introduced to decompose the network traffic and get several components. Approximate entropy is introduced to judge the complexity of each component. According to the results of approximate entropy, echo state network is selected to predict high complexity components, support vector machine is used to predict medium complexity components, and autoregressive integrated moving average model is introduced to predict low complexity components. The advantages of each prediction model are used to predict the appropriate component. The final prediction results are obtained by adding the predicted values of each component. In order to solve the problem that the prediction performance of support vector machine and echo state network is affected by their parameters, an improved whale optimization algorithm is proposed to optimize the parameters of the model. Meanwhile, the calculation results of approximate entropy show that compared with the original network traffic, the complexity of each component obtained by ensemble empirical mode decomposition is reduced, which reduces the complexity of modeling. Three network traffic datasets with sampling periods of 10 ms, 1 s, and 10 min are collected. Compared with the other five state‐of‐the‐art prediction models, the case study results show that the proposed prediction model has better prediction accuracy and excellent statistical performance indicators.

BackgroundThis study aimed to screen the feature modules and characteristic genes related to ulcerative colitis (UC) and construct a support vector machine (SVM) classifier to distinguish UC patients.MethodsFour datasets that contained UC and control samples were obtained from the Gene Expression Omnibus database. Differentially expressed genes (DEGs) with consistency were screened via the MetaDE method. The weighted gene coexpression network (WGCNA) was used to distinguish significant modules based on the four datasets. The protein–protein interaction network was established based on intersection genes. Enrichment analysis of Gene Ontology (GO) biological processes (BPs) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment were established based on DAVID. An SVM combined with recursive feature elimination was also applied to construct a disease classifier for the disease diagnosis of UC patients. The efficacy of the SVM classifier was evaluated through receiver operating characteristic curves.ResultsTwelve highly preserved modules were obtained using the WGCNA, and 2009 DEGs with significant consistency were selected using the MetaDE method. Sixteen significantly related GO BPs and 12 KEGG pathways were obtained, such as cytokine-cytokine receptor interaction, cell adhesion molecules, and leukocyte transendothelial migration. Subsequently, 41 genes were used to construct an SVM classifier, such as CXCL1, CCR2, IL1B, and IL1A. The area under the curve (AUC) was 0.999 in the training dataset, whereas the AUC was 0.886, 0.790, and 0.819 in the validation set (GSE65114, GSE37283, and GSE36807, respectively).ConclusionsAn SVM classifier based on feature genes might correctly identify healthy people or UC patients.

Aiming at the problem that the detection accuracy of effluent COD (chemical oxygen demand) in sewage treatment needs to be further improved, a combined model based on support vector machine and neural network is proposed to predict effluent COD. It can reduce the influence of local optimum on the global scope so as to improve the accuracy of prediction. Firstly, the sample data are divided into two categories by support vector machine. Then the BP neural network model and the Echo State Network (ESN) model are established on two sub-samples respectively. Compared with single neural network model, the mean absolute error and root mean square error of combined model are both reduced. Besides, the proposed model has better comprehensive prediction performance and can meet the actual demand of effluent COD prediction in sewage treatment.

Microcalcification (MC) clusters in mammograms can be important early signs of breast cancer in women. Accurate detection of MC clusters is an important but challenging problem. In this paper, we propose the use of a recently developed machine learning technique -- relevance vector machine (RVM) -- for automatic detection of MCs in digitized mammograms. RVM is based on Bayesian estimation theory, and as a feature it can yield a decision function that depends on only a very small number of so-called <i>relevance vectors</i>. We formulate MC detection as a supervised-learning problem, and use RVM to classify if an MC object is present or not at each location in a mammogram image. MC clusters are then identified by grouping the detected MC objects. The proposed method is tested using a database of 141 clinical mammograms, and compared with a support vector machine (SVM) classifier which we developed previously. The detection performance is evaluated using the free-response receiver operating characteristic (FROC) curves. It is demonstrated that the RVM classifier matches closely with the SVM classifier in detection performance, and does so with a much sparser kernel representation than the SVM classifier. Consequently, the RVM classifier greatly reduces the computational complexity, making it more suitable for real-time processing of MC clusters in mammograms.

e14528 Background: Treatment monitoring in patients with High-Grade Glioma (HGG) and identification of disease progression, remains a major challenge in clinical neurooncology. We aimed to develop a support vector machine (SVM) classifier utilising combined longitudinal conventional and Dynamic Susceptibility Contrast (DSC) perfusion MRI to classify between Stable Disease (SD), Pseudoprogression (PsP) and Progressive Disease (PD) in glioma patients under surveillance. Methods: Conventional (269) and perfusion (62) MRI studies of HGG patients acquired between 2012 and 2018 were prospectively analysed. Study participants were separated into two groups: Group I with a single DSC time point (64 participants) and Group II with multiple DSC time points (19 participants). The SVM classifier was trained using all available MRI for each group. Classification accuracy was assessed for the use of features extracted from different feature dataset and time point combinations and compared to the experienced radiologists’ predictions. Results: The study included 64 participants (mean age: 48.5 ± 12.8 yrs [standard deviation], 24 female). SVM classification based on combined perfusion and structural features outperformed standalone datasets across all groups. For the clinically relevant classification step (SD/PSP vs PD), both feature combination as well as the addition of multiple DSC time points, improved classification performance (lowest median error rate: 0.016). The SVM algorithm outperformed radiologists in predicting lesion destiny in both groups. Optimal performance was observed in Group II, in which SVM sensitivity/specificity/accuracy was 100/91.67/94.7% for analysis based on the first time point and 85.71/100/ 94.7% based on multiple time points compared to 60/78/68% and 70/90/84.2% for the respective radiologist classifications. In Group I, the SVM also exceeded radiologist predictions, albeit by a smaller margin and resulted in sensitivity/specificity of 86.49/75.00/81.53% (SVM) compared to 75.7/68.9/73.84% (radiologists). Conclusions: Our results indicate that the addition of multiple longitudinal perfusion time points as well as the combination of structural and perfusion features significantly enhance classification outcome in treatment monitoring of HGGs and machine-learning-assisted diagnosis has potentially superior accuracy than the radiologist's visual evaluation and expertise.

Sign language, also known as the silent language, is a vital communication tool for the hearing-impaired society where the communication represents a significant challenge. In this paper, we propose a gesture recognition system implementation for both static and dynamic gesture (Jochen Triesch’s static dataset for American Sign Language ASL, and Sebastien Marce’s dynamic database) in a uniform background. The system is based on Gabor features, Histogram of Oriented Gradients (HOG), Center-Symmetric Local Binary Patterns (CSLBP), Local Directional ZigZag Pattern (LDZP) and Local Energy based Shape Histogram (LESH) where the recognition task is applied using the one against all Support Vector Machine (SVM) classification technique. Individual textural and combined Gabor-textural descriptors are also studied to show the efficiency of the oriented Gabor filters. The combined Gabor-HOG architecture achieved the best recognition rate where 98.33 percent was obtained for ASL.

Failure of power transformer is very complex, so that it is difficult to use the mathematical model to describe their faults. In this study, an intelligent diagnostic method based on ant colony-support vector machine (AC-SVM) approach is presented for fault diagnosis of power transformer. The AC-SVM selects kernel function parameter and soft margin constant C penalty parameter of support vector machine (SVM) classifier. The performance of the AC-SVM system proposed in this study is evaluated by cases in China. The test results show that this AC-SVM model is effective to detect failure of power transformer.

The problem with development of the support vector machine (SVM) classifiers using modified particle swarm optimization (PSO) algorithm and their ensembles has been considered. Solving this problem would allow fulfilling the high-precision data classification, especially Big Data classification, with the acceptable time expenditures. The modified PSO algorithm conducts a simultaneous search of the type of kernel functions, the parameters of the kernel function and the value of the regularization parameter for the SVM classifier. The idea of particles' «regeneration» served as the basis for the modified PSO algorithm. In the implementation of this algorithm, some particles change the type of their kernel function to the one which corresponds to the particle with the best value of the classification accuracy. The offered PSO algorithm allows reducing the time expenditures for the developed SVM classifiers, which is very important for Big Data classification problem. In most cases such SVM classifier provides the high quality of data classification. In exceptional cases the SVM ensembles based on the decorrelation maximization algorithm for the different strategies of the decision-making on the data classification and the majority vote rule can be used. Also, the two-level SVM classifier has been offered. This classifier works as the group of the SVM classifiers at the first level and as the SVM classifier on the base of the modified PSO algorithm at the second level. The results of experimental studies confirm the efficiency of the offered approaches for Big Data classification.

Registration and difference analysis of corresponding mammogram images

Image classification is typically performed with highly trained feed-forward machine learning algorithms like deep neural networks and support vector machines. The image can be treated as a time-series input when applied to the network multiple times, opening the way for recurrent neural networks to perform tasks like image classification, semantic segmentation and auto-encoding. With this approach, ultra-fast training, network optimization, and short-term memory effects allows for dynamic, low-volume datasets to be quickly learned without heavy image pre-processing or feature extraction; the main limitation being that input images need labeled output images for training, as is also true of most standard approaches. In this work, the MNIST handwritten digit dataset is used as a benchmark to evaluate metrics of a modified Echo State Network for static image classification. The image array is passed through a noise filter multiple times as the Echo State Network converges to a classification. This highly dynamic approach easily adapts to sequential image (video) tasks like object tracking and is effective with small datasets. Classification rates reach 95.3% with sample size of 10000 handwritten digits and training time of approximately 5 minutes. Progression of this research enables discrete image and time-series classification under a single algorithm, with low computing power and memory requirements.