Search

The construction of cell differentiation trajectories from single-cell transcriptomic data or proteomic data by single-cell trajectory inference methods can help to understand the developmental process of normal tissues or provide pathologically relevant information. However, the accuracy and robustness of current single-cell trajectory inference algorithms are still a challenge, one of the reasons is the noise caused by the detection of a large number of unrelated genes in single-cell sequencing. In order to solve this problem, a trajectory inference method iterTIPD (iterative trajectory inference based on probability distribution) based on iterative feature selection is proposed. Its innovation lies in iteratively applying feature selection methods widely used for screening differentially expressed genes to linear or branching single-cell RNA sequencing data, and improving the accuracy and robustness of cell pseudotime ordering by selecting the gene subset that contributes the most to the constructed differentiation trajectory. Experimental results on four scRNA-seq datasets show that iterTIPD can effectively improve the accuracy and robustness of the single-cell trajectory inference algorithm. IterTIPD also improves the performance of other trajectory inference algorithms, proving generalization of iterTIPD. The differentiation trajectory of neural stem cells is reconstructed by iterTIPD algorithm, and the comparison shows that the differentiation trajectory is highly consistent with the known neural stem cell differentiation trajectory. Meanwhile, Top2a and Gja1 may be novel markers defining activated neural stem cell subpopulations.

Haixia Liu,1 Guozhong Cui,2 Yi Luo,3 Yajie Guo,1 Lianli Zhao,4 Yueheng Wang,5 Abdulhamit Subasi,6,7 Sengul Dogan,8 Turker Tuncer8 1Department of Ultrasound, Cangzhou Central Hospital, Cangzhou, Hebei Province, 061000, People’s Republic of China; 2Department of Surgical Oncology, Cangzhou Central Hospital, Cangzhou, Hebei Province, 061000, People’s Republic of China; 3Medical Statistics Room, Cangzhou Central Hospital, Cangzhou, Hebei Province, 061000, People’s Republic of China; 4Department of Internal Medicine teaching and research group, Cangzhou Central Hospital, Cangzhou, Hebei Province, 061000, China; 5Department of Ultrasound, The Second Hospital of Hebei MedicalUniversity, Shijiazhuang, Hebei Province, 050000, People’s Republic of China; 6Institute of Biomedicine, Faculty of Medicine, University of Turku, Turku, 20520, Finland; 7Department of Computer Science, College of Engineering, Effat University, Jeddah, 21478, Saudi Arabia; 8Department of Digital Forensics Engineering, College of Technology, Firat University, Elazig, 23119, TurkeyCorrespondence: Lianli Zhao, Department of Internal Medicine Teaching and Research Group, Cangzhou Central Hospital, Cangzhou, Hebei Province, 061000, People’s Republic of China, Tel +86 183 31125851, Email Zhaolianli123456@163.comPurpose: Breast cancer is a prominent cancer type with high mortality. Early detection of breast cancer could serve to improve clinical outcomes. Ultrasonography is a digital imaging technique used to differentiate benign and malignant tumors. Several artificial intelligence techniques have been suggested in the literature for breast cancer detection using breast ultrasonography (BUS). Nowadays, particularly deep learning methods have been applied to biomedical images to achieve high classification performances.Patients and Methods: This work presents a new deep feature generation technique for breast cancer detection using BUS images. The widely known 16 pre-trained CNN models have been used in this framework as feature generators. In the feature generation phase, the used input image is divided into rows and columns, and these deep feature generators (pre-trained models) have applied to each row and column. Therefore, this method is called a grid-based deep feature generator. The proposed grid-based deep feature generator can calculate the error value of each deep feature generator, and then it selects the best three feature vectors as a final feature vector. In the feature selection phase, iterative neighborhood component analysis (INCA) chooses 980 features as an optimal number of features. Finally, these features are classified by using a deep neural network (DNN).Results: The developed grid-based deep feature generation-based image classification model reached 97.18% classification accuracy on the ultrasonic images for three classes, namely malignant, benign, and normal.Conclusion: The findings obviously denoted that the proposed grid deep feature generator and INCA-based feature selection model successfully classified breast ultrasonic images.Keywords: deep classification framework, deep neural network, grid-based deep feature generator, iterative feature selection, breast ultrasonography (BUS)

Optical coherence tomography (OCT) images coupled with many learning techniques have been developed to diagnose retinal disorders. This work aims to develop a novel framework for extracting deep features from 18 pre-trained convolutional neural networks (CNN) and to attain high performance using OCT images. In this work, we have developed a new framework for automated detection of retinal disorders using transfer learning. This model consists of three phases: deep fused and multilevel feature extraction, using 18 pre-trained networks and tent maximal pooling, feature selection with ReliefF, and classification using the optimized classifier. The novelty of this proposed framework is the feature generation using widely used CNNs and to select the most suitable features for classification. The extracted features using our proposed intelligent feature extractor are fed to iterative ReliefF (IRF) to automatically select the best feature vector. The quadratic support vector machine (QSVM) is utilized as a classifier in this work. We have developed our model using two public OCT image datasets, and they are named database 1 (DB1) and database 2 (DB2). The proposed framework can attain 97.40% and 100% classification accuracies using the two OCT datasets, DB1 and DB2, respectively. These results illustrate the success of our model.

• A novel dynamic tree-based feature extraction function was proposed. • A new audio signal dataset was collected to detect violence. • A record-based validation technique (LORO CV) was presented. • Hand-modeled audio classification model was recommended and it achieved 89.75% accuracy. • Our proposal outperformed. Physical violence detection using multimedia data is crucial for public safety and security. This is an important research area in information security and digital forensics. Research in video-based violence detection (VVD) has grown steadily in recent years with rapid increase in video surveillance systems worldwide. Verbal aggression detection technologies, on the other hand, are still limited due to the popularity of computer vision models. Thus, researchers have preferred to use computer vision models to detect violence using videos. We have presented a new automatic audio violence detection (AVD) model to fill this gap. Our AVD model is handcrafted and its details are as follows. This work collected a new audio dataset on verbal aggression from YouTube. A novel handcrafted model was proposed using multilevel feature extraction, feature selection, classification, and majority voting phases. A new local feature extraction function based on the binary tree was used to generate features from audio signals. We call this function tree pattern-23 (TreePat23), where 23 represents the number of wavelet bands/audio signals inputs. Wavelet bands were generated using tunable Q wavelet transform (TQWT) before being applied to our TreePat23 for feature extraction. The iterative neighborhood component analysis (INCA) and Chi2 were used to select the features. The selected features were classified using k nearest neighbors (kNN) and support vector machine (SVM) followed by iterative majority voting (IMV) method. The best-predicted vector was obtained by using a greedy algorithm. Finally, a new validation technique called leave one record out (LORO) cross-validation (CV) was used to validate the results. Our proposed TreePat23 model has attained classification accuracy of 89.68% and 89.75% with kNN and SVM, respectively. Our developed system has generated 14 results for each classifier and automatically selected the best result. Hence this model is a self-organized audio classification model which yielded over 89% classification accuracy for both classifiers using LORO CV strategy. [ABSTRACT FROM AUTHOR]

This paper presents a failure prediction model using iterative feature selection, which aims to accurately predict the failure occurrences in industrial Internet of Things (IIoT) environments. In general, vast amounts of data are collected from various sensors in an IIoT environment, and they are analyzed to prevent failures by predicting their occurrence. However, the collected data may include data irrelevant to failures and thereby decrease the prediction accuracy. To address this problem, we propose a failure prediction model using iterative feature selection. To build the model, the relevancy between each feature (i.e., each sensor) and the failure was analyzed using the random forest algorithm, to obtain the importance of the features. Then, feature selection and model building were conducted iteratively. In each iteration, a new feature was selected considering the importance and added to the selected feature set. The failure prediction model was built for each iteration via the support vector machine (SVM). Finally, the failure prediction model having the highest prediction accuracy was selected. The experimental implementation was conducted using open-source R. The results showed that the proposed failure prediction model achieved high prediction accuracy.

Ensuring the security of critical infrastructures and information systems has always been a challenging problem. Intrusion Detection Systems (IDS) have become an indispensable element in ensuring the security of network structures and information systems. it is important to increase the intrusion detection performance of IDS systems. Although a lot of work has been done to improve IDS systems, there are still many shortcomings. In this study, a new Switch port anomaly based IDS (SPA-IDS) dataset is proposed for IDS systems. The data is taken over switch ports with normal internet traffic. This research presents a new automated intrusion classification model using the collected signals. These signals acquired from layer 2. The presented automated classification model consists of feature generation using vertical mode decomposition (VMD) and statistics, iterative feature selection and classification phases. These phases are given as follows. (i) VMD is applied to the acquired signals and five VMD coefficients are obtained. 15 statistical moments are applied to signal and the calculated VMD coefficients. (ii) Iterative feature selection is applied to extracted features and the most valuable features are selected. (iii) The chosen features are classified using decision tree (DT), bagged tree (BT), support vector machine (SVM) and k nearest neighbor (KNN) machine learning methods with ten-fold cross validation. The presented VMD and iterative feature selection based model attained 87.25%, 95.98%, 96.51% and 97.74% accuracies by employing KNN, SVM, DT and BT classifiers consecutively. After the attack categorization process in the second phase of the study, the same classifiers success rates reached 96.65%, 98.52%, 98.39% and 99.11%, respectively. The calculated/obtained accuracies obviously denotes the success of the presented VMD and iterative feature selection based intrusion detection system. Owing to the method presented in this study, we propose an effective and fast IDS approach by analyzing the packets received at layer-2 in order to prevent attacks from the network. [ABSTRACT FROM AUTHOR]

• New MR image datasets was collected to diagnose two shoulder ailments. • A patch-based VGG19 has been presented to extract features. • A new image classification model (ViVGG19) has been presented using our feature generator. • Three shallow classifiers have been used to denote high classification ability ViVGG19. • The ViVGG19 attained over 99.5% classification accuracies for the used all datasets. : Rotator cuff tear (RCT) and biceps tendinosis (BT) are the two most common shoulder disorders worldwide. These disorders can be diagnosed using magnetic resonance imaging (MRI), but the expert interpretation is manual, time-consuming, and subjected to human errors. Therefore, a fixed-size feature extraction model was created to objectively and accurately perform automated binary classification of RCT vs. normal and BT vs. normal on MRI images. : We have developed an exemplar deep feature extraction model to diagnose RCT and BT disorders. The model was tested on a new MR image dataset comprising transverse, sagittal, and coronal MRI images of the shoulder that had been organized into three cases. BT was studied on transverse MRI images (Case 1), while RCT was studied on sagittal (Case 2) and coronal MRI images (Case 3). Our model comprised deep feature generation using a pre-trained VGG19, feature selection using iterative neighborhood component analysis (INCA), and classification using shallow standard classifiers k-nearest neighbors (KNN), support vector machine (SVM), and artificial neural network (ANN). In the feature extraction phase, two fully connected layers were used to extract deep features from the original image, and sixteen fixed-size patches obtained by the division of the original image. This model was named Vision VGG19 (ViVGG), analogous to vision transformers (ViT). The feature vector is extracted from the raw image dataset, and 16 feature vectors are extracted from each fixed-size patch. Seventeen feature vectors obtained from each image are obtained from fc6 and fc7 layers of the pre-trained VGG19, are merged to obtain final feature vector. INCA was used to choose the top features from the created features, and the chosen features were classified using shallow classifiers. : We defined three cases to evaluate the proposed ViVGG19 to diagnose RT and BCT disorders. Our proposed ViVGG19 model achieved more than 99% accuracy using the KNN classifier. : ViVGG19 is a very effective model for detecting RCT and BT disorders on shoulder MRI images. The developed automated system is ready to be tested with a bigger diverse database obtained from different medical centers. [ABSTRACT FROM AUTHOR]

Two important problems which can affect the performance of classification models are high-dimensionality (an overabundance of independent features in the dataset) and imbalanced data (a skewed class distribution which creates at least one class with many fewer instances than other classes). To resolve these problems concurrently, we propose an iterative feature selection approach, which repeated applies data sampling (in order to address class imbalance) followed by feature selection (in order to address high-dimensionality), and finally we perform an aggregation step which combines the ranked feature lists from the separate iterations of sampling. This approach is designed to find a ranked feature list which is particularly effective on the more balanced dataset resulting from sampling while minimizing the risk of losing data through the sampling step and missing important features. To demonstrate this technique, we employ 18 different feature selection algorithms and Random Undersampling with two post-sampling class distributions. We also investigate the use of sampling and feature selection without the iterative step (e.g., using the ranked list from a single iteration, rather than combining the lists from multiple iterations), and compare these results from the version which uses iteration. Our study is carried out using three groups of datasets with different levels of class balance, all of which were collected from a real-world software system. All of our experiments use four different learners and one feature subset size. We find that our proposed iterative feature selection approach outperforms the non-iterative approach. [ABSTRACT FROM AUTHOR]

Software quality prediction models are useful tools for creating high quality software products. The general process is that practitioners use software metrics and defect data along with various data mining techniques to build classification models for identifying potentially faulty program modules, thereby enabling effective project resource allo-cation. The predictive accuracy of these classification models is often affected by the quality of input data. Two main problems which can affect the quality of input data are high dimensionality (too many independent attributes in a dataset) and class imbal-ance (many more members of one class than the other class in a binary classification problem). To resolve both of these problems, we present an iterative feature selection approach which repeatedly applies data sampling (to overcome class imbalance) followed by feature selection (to overcome high dimensionality), and finally combines the ranked feature lists from the separate iterations of sampling. After feature selection, models are built either using a plain learner or by using a boosting algorithm which incorporates sampling. In order to assess the impact of various balancing, filter, and learning tech-niques in the feature selection and model-building process on software quality prediction, we employ two sampling techniques, random undersampling (RUS) and synthetic minor-ity oversampling technique (SMOTE), and two ensemble boosting approaches, RUSBoost and SMOTEBoost (in which RUS and SMOTE, respectively, are integrated into a boosting technique), as well as six feature ranking techniques. We apply the proposed techniques to several groups of datasets from two real-world software systems and use two learners to build classification models. The experimental results demonstrate that RUS results in better prediction than SMOTE, and also that boosting is more effective in improving classification performance than not using boosting. In addition, some feature ranking techniques, like chi-squared and information gain, exhibit better and more stable classi-fication behavior than other rankers. [ABSTRACT FROM AUTHOR]

Many diseases have a genetic origin, and a great effort is being made to detect the genes that are responsible for their insurgence. One of the most promising techniques is the analysis of genetic information through the use of complex networks theory. Yet, a practical problem of this approach is its computational cost, which scales as the square of the number of features included in the initial dataset. In this paper, we propose the use of an iterative feature selection strategy to identify reduced subsets of relevant features, and show an application to the analysis of congenital Obstructive Nephropathy. Results demonstrate that, besides achieving a drastic reduction of the computational cost, the topologies of the obtained networks still hold all the relevant information, and are thus able to fully characterize the severity of the disease. [ABSTRACT FROM AUTHOR]

Optical coherence tomography (OCT) images coupled with many learning techniques have been developed to diagnose retinal disorders. This work aims to develop a novel framework for extracting deep features from 18 pre-trained convolutional neural networks (CNN) and to attain high performance using OCT images. In this work, we have developed a new framework for automated detection of retinal disorders using transfer learning. This model consists of three phases: deep fused and multilevel feature extraction, using 18 pre-trained networks and tent maximal pooling, feature selection with ReliefF, and classification using the optimized classifier. The novelty of this proposed framework is the feature generation using widely used CNNs and to select the most suitable features for classification. The extracted features using our proposed intelligent feature extractor are fed to iterative ReliefF (IRF) to automatically select the best feature vector. The quadratic support vector machine (QSVM) is utilized as a classifier in this work. We have developed our model using two public OCT image datasets, and they are named database 1 (DB1) and database 2 (DB2). The proposed framework can attain 97.40% and 100% classification accuracies using the two OCT datasets, DB1 and DB2, respectively. These results illustrate the success of our model. [ABSTRACT FROM AUTHOR]

This paper presents a failure prediction model using iterative feature selection, which aims to accurately predict the failure occurrences in industrial Internet of Things (IIoT) environments. In general, vast amounts of data are collected from various sensors in an IIoT environment, and they are analyzed to prevent failures by predicting their occurrence. However, the collected data may include data irrelevant to failures and thereby decrease the prediction accuracy. To address this problem, we propose a failure prediction model using iterative feature selection. To build the model, the relevancy between each feature (i.e., each sensor) and the failure was analyzed using the random forest algorithm, to obtain the importance of the features. Then, feature selection and model building were conducted iteratively. In each iteration, a new feature was selected considering the importance and added to the selected feature set. The failure prediction model was built for each iteration via the support vector machine (SVM). Finally, the failure prediction model having the highest prediction accuracy was selected. The experimental implementation was conducted using open-source R. The results showed that the proposed failure prediction model achieved high prediction accuracy. [ABSTRACT FROM AUTHOR]

Background & Aims: Almost all gastric cancers are adenocarcinomas, which have considerable heterogeneity among patients. We sought to identify subtypes of gastric adenocarcinomas with particular biological properties and responses to chemotherapy and targeted agents., Methods: We compared gene expression patterns among 248 gastric tumors; using a robust method of unsupervised clustering, consensus hierarchical clustering with iterative feature selection, we identified 3 major subtypes. We developed a classifier for these subtypes and validated it in 70 tumors from a different population. We identified distinct genomic and epigenomic properties of the subtypes. We determined drug sensitivities of the subtypes in primary tumors using clinical survival data, and in cell lines through high-throughput drug screening., Results: We identified 3 subtypes of gastric adenocarcinoma: proliferative, metabolic, and mesenchymal. Tumors of the proliferative subtype had high levels of genomic instability, TP53 mutations, and DNA hypomethylation. Cancer cells of the metabolic subtype were more sensitive to 5-fluorouracil than the other subtypes. Furthermore, in 2 independent groups of patients, those with tumors of the metabolic subtype appeared to have greater benefits with 5-fluorouracil treatment. Tumors of the mesenchymal subtype contain cells with features of cancer stem cells, and cell lines of this subtype are particularly sensitive to phosphatidylinositol 3-kinase-AKT-mTOR inhibitors in vitro., Conclusions: Based on gene expression patterns, we classified gastric cancers into 3 subtypes, and validated these in an independent set of tumors. The subgroups have differences in molecular and genetic features and response to therapy; this information might be used to select specific treatment approaches for patients with gastric cancer., (Copyright © 2013 AGA Institute. Published by Elsevier Inc. All rights reserved.)