Recursive Feature Extraction Research Articles

A comparative study of ensemble methods and multi-output classifiers for predictive maintenance of hydraulic systems

The maintenance and upkeep of hydraulic systems play a pivotal role in various industrial applications, from manufacturing processes to heavy machinery operations. Traditional strategies often face significant limitations, particularly concerning intervention time and the costs associated with delays between fault occurrence and maintenance. Over the past decades, proactive strategies have emerged with promising potential, primarily due to their predictive capabilities. These proactive approaches aim to anticipate faults and maintenance needs, thereby mitigating costs and operational disruptions associated with reactive approaches. This study explores the applicability of single output, ensemble methods, and the integration of multi-output classifiers in hydraulic predictive maintenance problems with relatively limited data. First, data is analyzed using Pearson correlation coefficients, and then feature extraction is conducted using recursive feature extraction, aiming to optimize the performance of predictive models, particularly Random Forest and CatBoost. Results show that the stacking ensemble method, incorporating LightGBM, XGBoost, CatBoost, and Random Forest, yields the most notable improvement, achieving a final accuracy of 98.63%. The results obtained in this study show satisfying performances for single-output models, ensemble methods, and multi-output models in predicting the health of hydraulic systems. Moreover, combining single output, ensemble methods, and the integration of multi-output classifiers has created a relatively reliable and robust predictive maintenance system.

Performance Enhancement in Machine Learning Approaches using Recursive Feature Extraction, Maximum Variance Method and Min Max Method

Performance Enhancement in Machine Learning Approaches using Recursive Feature Extraction, Maximum Variance Method and Min Max Method

Precision healthcare: A deep dive into machine learning algorithms and feature selection strategies for accurate heart disease prediction

This paper presents a comprehensive exploration of machine learning algorithms (MLAs) and feature selection techniques for accurate heart disease prediction (HDP) in modern healthcare. By focusing on diverse datasets encompassing various challenges, the research sheds light on optimal strategies for early detection. MLAs such as Decision Trees (DT), Random Forests (RF), Support Vector Machines (SVM), Gaussian Naive Bayes (NB), and others were studied, with precision and recall metrics emphasized for robust predictions. Our study addresses challenges in real-world data through data cleaning and one-hot encoding, enhancing the integrity of our predictive models. Feature extraction techniques—Recursive Feature Extraction (RFE), Principal Component Analysis (PCA), and univariate feature selection—play a crucial role in identifying relevant features and reducing data dimensionality. Our findings showcase the impact of these techniques on improving prediction accuracy. Optimized models for each dataset have been achieved through grid search hyperparameter tuning, with configurations meticulously outlined. Notably, a remarkable 99.12 % accuracy was achieved on the first Kaggle dataset, showcasing the potential for accurate HDP. Model robustness across diverse datasets was highlighted, with caution against overfitting. The study emphasizes the need for validation of unseen data and encourages ongoing research for generalizability. Serving as a practical guide, this research aids researchers and practitioners in HDP model development, influencing clinical decisions and healthcare resource allocation. By providing insights into effective algorithms and techniques, the paper contributes to reducing heart disease-related morbidity and mortality, supporting the healthcare community's ongoing efforts.

Predicting Progression of Intracranial Hemorrhage in the Prehospital TXA for TBI Trial.

Progression of intracranial hemorrhage is a common, potentially devastating complication after moderate/severe traumatic brain injury (TBI). Clinicians have few tools to predict which patients with traumatic intracranial hemorrhage on their initial head computed tomography (hCT) scan will progress. The objective of this investigation was to identify clinical, imaging, and/or protein biomarkers associated with progression of intracranial hemorrhage (PICH) after moderate/severe TBI and to create an accurate predictive model of PICH based on clinical features available at presentation. We analyzed a subset of subjects from the phase II double-blind, multi-center, randomized "Prehospital Tranexamic Acid Use for TBI" trial. This subset was limited to the placebo arm of the parent trial with evidence of hemorrhage on the initial hCT and a follow-up hCT 6 h after. PICH was defined as an increase in hemorrhage size by 30% or more, or the development of new hemorrhage in the intra- and extra-axial intracranial vault between the initial and the follow-up hCT. Two independent radiologists evaluated each hCT, and conflicts were adjudicated by a third. Clinical and radiographic characteristics were collected, along with plasma protein biomarkers at admission. Principal component analysis (PCA) was performed, and each principal component (PC) was interrogated for its association with PICH. Finally, expert opinion and recursive feature extraction (RFE) were used to select input features for the construction of several supervised classification models. Their ability to predict PICH was quantified and compared. In this subset of subjects (n = 104), 46% (n = 48) demonstrated PICH. Univariate analyses showed no association between PICH and age, sex, admission Glasgow Coma Scale (GCS), GCS motor subscore, presence of midline shift, admission platelet count or admission INR. Radiographic severity scores (Marshall score [p = 0.007], Rotterdam score [p = 0.004]), and initial hematoma volume [p = 0.005] were associated with PICH. Higher levels of admission glial fibrillary acidic protein (p < 0.001) and MAP (p = 0.011) were also associated with PICH. Of the PCs, PC1 was significantly associated with PICH (p = 0.0125). Using multimodal data input, machine learning classifiers successfully discriminated patients with or without PICH. Models composed of machine-selected features performed better than models composed of expert-selected variables (reaching an average of 77% accuracy, AUC = 0.78 versus AUC = 0.68 for the expert-selected variables). Predictive models utilizing variables measured at admission can accurately predict PICH, confirmed by the 6-hour follow-up hCT. Our best-performing models must now be externally validated in a separate cohort of TBI patients with low GCS and initial hCT positive for hemorrhage.

Machine Learning Approach to Identify Dysregulation in the Plasma Proteome of CADASIL

AbstractBackgroundThe plasma proteome provides information to identify dysregulated molecular pathways underlying disease. Cerebrovascular diseases often manifest a significantly different plasma proteomic signature. Cerebral‐Autosomal‐Dominant‐Arteriopathy‐Subcortical‐Infarcts‐and‐Leukoencephalopathy (CADASIL) is one such vascular brain disease. The CADASIL plasma proteome has yet to be investigated.MethodTo investigate the plasma proteome in the context of CADASIL, we used large‐scale proteomics, measuring over 7,000 proteins via aptamer‐based technology (SomaLogic) from 53 study participants (nCADASIL = 25; nControl = 28). We employed machine learning (ML) methods as an unbiased approach to uncover disease‐associated change in proteomic networks. This approach is based on the premise that protein sets that best classify disease state could be important biological drivers of disease. We developed a novel ML method: coupling recursive feature extraction with logistic regression (LR) and XGBoost evaluators, as well maximum‐relevance‐minimum‐redundancy with Random‐Forest and F‐Statistic evaluators. The results of all four models were selectively aggregated in order to delineate the CADASIL plasma proteome signature while minimizing overfitting to the low sample size data. We developed a 45 protein model of the CADASIL proteomic signature from these results. To evaluate the classification ability, we tested the Repeated‐Stratified‐10‐fold‐Cross‐Validation accuracy of a LR classifier. Next, we investigated whether the CADASIL proteomic signature was relevant for other neurodegenerative diseases with vascular components; the classifier was applied to an Alzheimer’s Disease (AD) dataset. Finally, the functional signature of the 45 proteins was investigated through gene ontology (GO).ResultThe LR classifier in CADASIL is 99.8% accurate. In AD, the accuracy is 71.2%. This implies the CADASIL signature shares commonalities as well as pertinent differences with other neurodegenerative diseases. The GO of the 45 proteins resulted in GO terms corresponding to extracellular matrix and collagen, cellular membrane lipoproteins and glycolipids, and proteasome biological pathways.ConclusionThe proposed method allows for an initial unbiased discovery of important proteomic changes associated with disease state. The identified proteins would provide starting points for mechanistic studies. In addition, a panel of proteins could be used for screening at population level for risk stratification and more in‐depth phenotyping. Next steps include investigating the biological relevance of findings in mechanistic models for therapeutic developments.

Machine Learning Approach to Identify Dysregulation in the Plasma Proteome of CADASIL

AbstractBackgroundThe plasma proteome provides information to identify dysregulated molecular pathways underlying disease. Cerebrovascular diseases often manifest a significantly different plasma proteomic signature. Cerebral‐Autosomal‐Dominant‐Arteriopathy‐Subcortical‐Infarcts‐and‐Leukoencephalopathy (CADASIL) is one such vascular brain disease. The CADASIL plasma proteome has yet to be investigated.MethodTo investigate the plasma proteome in the context of CADASIL, we used large‐scale proteomics, measuring over 7,000 proteins via aptamer‐based technology (SomaLogic) from 53 study participants (nCADASIL = 25; nControl = 28).We employed machine learning (ML) methods as an unbiased approach to uncover disease‐associated change in proteomic networks. This approach is based on the premise that protein sets that best classify disease state could be important biological drivers of disease. We developed a novel ML method: coupling recursive feature extraction with logistic regression (LR) and XGBoost evaluators, as well maximum‐relevance‐minimum‐redundancy with Random‐Forest and F‐Statistic evaluators.The results of all four models were selectively aggregated in order to delineate the CADASIL plasma proteome signature while minimizing overfitting to the low sample size data. We developed a 45 protein model of the CADASIL proteomic signature from these results. To evaluate the classification ability, we tested the Repeated‐Stratified‐10‐fold‐Cross‐Validation accuracy of a LR classifier. Next, we investigated whether the CADASIL proteomic signature was relevant for other neurodegenerative diseases with vascular components; the classifier was applied to an Alzheimer’s Disease (AD) dataset. Finally, the functional signature of the 45 proteins was investigated through gene ontology (GO).ResultsThe LR classifier in CADASIL is 99.8% accurate. In AD, the accuracy is 71.2%. This implies the CADASIL signature shares commonalities as well as pertinent differences with other neurodegenerative diseases.The GO of the 45 proteins resulted in GO terms corresponding to extracellular matrix and collagen, cellular membrane lipoproteins and glycolipids, and proteasome biological pathways.ConclusionThe proposed method allows for an initial unbiased discovery of important proteomic changes associated with disease state. The identified proteins would provide starting points for mechanistic studies. In addition, a panel of proteins could be used for screening at population level for risk stratification and more in‐depth phenotyping. Next steps include investigating the biological relevance of findings in mechanistic models for therapeutic developments.

Predicting Return to Work after Cardiac Rehabilitation.

To explore machine learning models for predicting return to work after cardiac rehabilitation. Patients who were admitted to the University of Malaya Medical Centre due to cardiac events. Eight different machine learning models were evaluated. The models included 3 different sets of features: full features; significant features from multiple logistic regression; and features selected from recursive feature extraction technique. The performance of the prediction models with each set of features was compared. The AdaBoost model with the top 20 features obtained the highest performance score of 92.4% (area under the curve; AUC) compared with other prediction models.

Untargeted stable isotope-resolved metabolomics to assess the effect of PI3Kβ inhibition on metabolic pathway activities in a PTEN null breast cancer cell line.

The combination of high-resolution LC-MS untargeted metabolomics with stable isotope-resolved tracing is a promising approach for the global exploration of metabolic pathway activities. In our established workflow we combine targeted isotopologue feature extraction with the non-targeted X13CMS routine. Metabolites, detected by X13CMS as differentially labeled between two biological conditions are subsequently integrated into the original targeted library. This strategy enables monitoring of changes in known pathways as well as the discovery of hitherto unknown metabolic alterations. Here, we demonstrate this workflow in a PTEN (phosphatase and tensin homolog) null breast cancer cell line (MDA-MB-468) exploring metabolic pathway activities in the absence and presence of the selective PI3Kβ inhibitor AZD8186. Cells were fed with [U-13C] glucose and treated for 1, 3, 6, and 24 h with 0.5 µM AZD8186 or vehicle, extracted by an optimized sample preparation protocol and analyzed by LC-QTOF-MS. Untargeted differential tracing of labels revealed 286 isotope-enriched features that were significantly altered between control and treatment conditions, of which 19 features could be attributed to known compounds from targeted pathways. Other 11 features were unambiguously identified based on data-dependent MS/MS spectra and reference substances. Notably, only a minority of the significantly altered features (11 and 16, respectively) were identified when preprocessing of the same data set (treatment vs. control in 24 h unlabeled samples) was performed with tools commonly used for label-free (i.e. w/o isotopic tracer) non-targeted metabolomics experiments (Profinder´s batch recursive feature extraction and XCMS). The structurally identified metabolites were integrated into the existing targeted isotopologue feature extraction workflow to enable natural abundance correction, evaluation of assay performance and assessment of drug-induced changes in pathway activities. Label incorporation was highly reproducible for the majority of isotopologues in technical replicates with a RSD below 10%. Furthermore, inter-day repeatability of a second label experiment showed strong correlation (Pearson R 2 > 0.99) between tracer incorporation on different days. Finally, we could identify prominent pathway activity alterations upon PI3Kβ inhibition. Besides pathways in central metabolism, known to be changed our workflow revealed additional pathways, like pyrimidine metabolism or hexosamine pathway. All pathways identified represent key metabolic processes associated with cancer metabolism and therapy.

A statistical focus on doping using a metabolomics approach

Identify potential biomarkers that are indicative of doping in equine plasma through statistical analysis of metabolomics data. The purpose of metabolomics is to detect endogenous changes in molecular entities, such as metabolites and adducts, that are indicative of challenges to the system (Fiehn et al., Metabolomics, 2015, 11, 1036–1040). These challenges to the system may include things such as doping, disease and environmental changes. Doping is an ever-growing field due to the changing nature of new and popular agents. For the racing industry, any performance altering agents are prohibited to maintain the welfare of athletes involved and the integrity of the sport and breeding industry (Cawley et al., Drug Testing and Analysis, 2017, 9, 1441–1447). Therefore, an alternative approach to conventional detection methods is essential for the decreasing the bottleneck that is the introduction of new compounds into routine screening efforts. Using only 100 μL of equine plasma, a rapid protein precipitation method was developed for the analysis of endogenous compounds. The IMTAKT Intrada Amino Acid column (100 mm × 2 mm, 3 μm) was able to separate dopamine-related compounds. The method used positive and negative ionisation mode with an 11-minute gradient method on the Agilent 1290 Infinity II LC system coupled to an Agilent 6545 QTOF mass spectrometer. A substantial reference population study was completed to assess basal concentrations of endogenous compounds of interest. A 12-horse administration study of Stalevo® (800 mg levodopa, 200 mg carbidopa, 1600 mg entacapone) was analysed. Agilent Technologies’ Profinder was used to complete a batch recursive feature extraction of the data. The statistical analysis included longitudinal profiling and identification of significant entities through volcano plot analysis, principal component analysis and heatmap visualisation. Biomarker quality was determined using a ‘best biomarker quality (BBQ)’ assessment. The LC-QTOF-MS method was successfully with respect to the quantification of 3-methoxytyrosine. The reference population study revealed variable levels of 3-methoxytyrosine in the population thus monitoring through individual reference limits was proposed. The longitudinal profiling approach, using 3-methoxytyrosine as an up-regulated biomarker, was proven to be more sensitive than a threshold proposed from a population. The untargeted approach using Profinder extracted a wealth of data across the administration study. The statistical analysis was able to determine significant biomarkers. The most significant biomarkers were investigated and attempted to be identified through database searching and confirmation with reference standards. An individualised reference limit approach using longitudinal profiling allowed for raceday control of levodopa misuse within the equine athlete. The untargeted metabolomics method provided an interesting alternative to conventional detection methods thus enabling the identification of biomarkers that are indicative of doping. Biomarker ratios were investigated to provide more statistical power than a single biomarker. Further work investigating an orthogonal approach using a reverse phase analytical method would allow for wider coverage of the metabolome. Doping continues to be a threat to the integrity of horse racing and thus new detection methods, such as metabolomics and statistical profiling, must be implemented to combat this challenge. 3-methoxytyrosine was identified to be a viable up-regulated biomarker for doping of levodopa. An untargeted metabolomics approach enabled the detection of further biomarkers thus allowing for a more statistically powerful biomarker ratio to be investigated.

Multiphasic CT-Based Radiomics Analysis for the Differentiation of Benign and Malignant Parotid Tumors.

ObjectiveThis study aims to investigate the value of machine learning models based on clinical-radiological features and multiphasic CT radiomics features in the differentiation of benign parotid tumors (BPTs) and malignant parotid tumors (MPTs).MethodsThis retrospective study included 312 patients (205 cases of BPTs and 107 cases of MPTs) who underwent multiphasic enhanced CT examinations, which were randomly divided into training (N = 218) and test (N = 94) sets. The radiomics features were extracted from the plain, arterial, and venous phases. The synthetic minority oversampling technique was used to balance minority class samples in the training set. Feature selection methods were done using the least absolute shrinkage and selection operator (LASSO), mutual information (MI), and recursive feature extraction (RFE). Two machine learning classifiers, support vector machine (SVM), and logistic regression (LR), were then combined in pairs with three feature selection methods to build different radiomics models. Meanwhile, the prediction performances of different radiomics models based on single phase (plain, arterial, and venous phase) and multiphase (three-phase combination) were compared to determine which model construction method and phase were more discriminative. In addition, clinical models based on clinical-radiological features and combined models integrating radiomics features and clinical-radiological features were established. The prediction performances of the different models were evaluated by the area under the receiver operating characteristic (ROC) curve (AUC) and the drawing of calibration curves.ResultsAmong the 24 established radiomics models composed of four different phases, three feature selection methods, and two machine learning classifiers, the LASSO-SVM model based on a three-phase combination had the optimal prediction performance with AUC (0.936 [95% CI = 0.866, 0.976]), sensitivity (0.78), specificity (0.90), and accuracy (0.86) in the test set, and its prediction performance was significantly better than with the clinical model based on LR (AUC = 0.781, p = 0.012). In the test set, the combined model based on LR had a lower AUC than the optimal radiomics model (AUC = 0.933 vs. 0.936), but no statistically significant difference (p = 0.888).ConclusionMultiphasic CT-based radiomics analysis showed a machine learning model based on clinical-radiological features and radiomics features has the potential to provide a valuable tool for discriminating benign from malignant parotid tumors.

Detection of DDoS Attack in IoT Networks Using Sample Selected RNN-ELM

The Internet of Things (IoT) is a global information and communication technology which aims to connect any type of device to the internet at any time and in any location. Nowadays billions of IoT devices are connected to the world, this leads to easily cause vulnerability to IoT devices. The increasing of users in different IoT-related applications leads to more data attacks is happening in the IoT networks after the fog layer. To detect and reduce the attacks the deep learning model is used. In this article, a hybrid sample selected recurrent neural network-extreme learning machine (hybrid SSRNN-ELM) algorithm that uses recurrent neural network (RNN) as a supervised and extreme learning machine (ELM) classifier as unsupervised. In the proposed algorithm sample selected features are extracting from the original dataset using linear regression with recursive feature extraction (LR-RFE) and sequence forward selector (SFS) then RNN is used to learn the behavior of the important features and at end layer the ELM classifier is used. This hybrid intrusion detection algorithm is placed in between the fog layer and its devices. NSL_KDD benchmark is used for detecting the distributed denial-of-service (DDoS) attack in IoT devices after the fog node. The proposed hybrid SSRNN-ELM model exposes the attacks while testing with enhanced accuracy of up to 99% from NSL-KDD data set. Experimental results outperform by using proposed technique when compared with the existing models.

Covid-19 and Pneumonia Infection Detection from Chest X-Ray Images using U-Net, EfficientNetB1, XGBoost and Recursive Feature Elimination

The pandemic caused by the COVID-19 virus is the most serious current threat to the public's health. For the purpose of identifying patients with Covid-19, Chest X-Rays have proven to be an indispensable imaging modality for the hospital. Nevertheless, radiologists are needed to commit a significant amount of time to their interpretation. It is possible to diagnose and triage cases of Covid-19 effectively and rapidly with the assistance of precise computer systems that are powered by Machine Learning techniques. Machine Learning techniques such as Deep Feature Extraction can help detect the disease with improved precision and speed when used in conjunction with X-Ray images of the lung. This helps to alleviate the problem of lack of testing kits. Using the U-Net for Semantic image segmentation for lung segmentation and deep feature extraction-based strategy that was suggested in this research, it is possible to differentiate between patients who have contracted the Covid-19 virus, pneumonia and healthy people. XGBoost and recursive feature extraction based proposed methodology is evaluated using 20 different Pre-Trained deep learning based models including EfficientNet variations and it is observed that the maximum detection accuracy, precision, recall specificity, and F1-score are achieved when EfficientNetB1 is used to extract deep features. The respective values for these metrics are 97.6%, 0.964, 0.964, and 0.982. These findings lend credence to the efficiency of the proposed methodology.

Machine learning approaches to predict the Plant-associated phenotype of Xanthomonas strains

BackgroundThe genus Xanthomonas has long been considered to consist predominantly of plant pathogens, but over the last decade there has been an increasing number of reports on non-pathogenic and endophytic members. As Xanthomonas species are prevalent pathogens on a wide variety of important crops around the world, there is a need to distinguish between these plant-associated phenotypes. To date a large number of Xanthomonas genomes have been sequenced, which enables the application of machine learning (ML) approaches on the genome content to predict this phenotype. Until now such approaches to the pathogenomics of Xanthomonas strains have been hampered by the fragmentation of information regarding pathogenicity of individual strains over many studies. Unification of this information into a single resource was therefore considered to be an essential step.ResultsMining of 39 papers considering both plant-associated phenotypes, allowed for a phenotypic classification of 578 Xanthomonas strains. For 65 plant-pathogenic and 53 non-pathogenic strains the corresponding genomes were available and de novo annotated for the presence of Pfam protein domains used as features to train and compare three ML classification algorithms; CART, Lasso and Random Forest.ConclusionThe literature resource in combination with recursive feature extraction used in the ML classification algorithms provided further insights into the virulence enabling factors, but also highlighted domains linked to traits not present in pathogenic strains.

Machine learning-based texture analysis for differentiation of radiologically indeterminate small adrenal tumors on adrenal protocol CT scans.

To evaluate the ability of radiomic feature extraction and a machine learning algorithm to differentiate between benign and malignant indeterminate adrenal lesions on contrast-enhanced computed tomography (CT) studies. Adrenal "incidentalomas" are adrenal lesions that are accidentally discovered during workup not related to the adrenal glands; they have an incidence as high as 5%. Small adrenal incidentalomas (< 4cm) with high attenuation values on pre-contrast CT(> 10 HU) need further evaluation to calculate the absolute percentage of washout (APW). If the APW is < 60%, these lesions are considered non-adenomas and commonly classified as indeterminate adrenal lesions. Further workup for indeterminate lesions includes more complicated and expensive radiological studies or invasive procedures like biopsy or surgical resection. We searched our institutional database for indeterminate adrenal lesions with the following characteristics: < 4cm, pre-attenuation value > 10 HU, and APW < 60%. Exclusion criteria included pheochromocytoma and no histopathological examination. CT images were converted to Nifti format, and adrenal tumors were segmented using Amira software. Radiomic features from the adrenal mask were extracted using PyRadiomics software after removing redundant features (highly pairwise correlated features and low-variance features) using recursive feature extraction to select the final discriminative set of features. Lastly, the final features were used to build a binary classification model using a random forest algorithm, which was validated and tested using leave-one-out cross-validation, confusion matrix, and receiver operating characteristic curve. We found 40 indeterminate adrenal lesions (21 benign and 19 malignant). Feature extraction resulted in 3947 features, which reduced down to 62 features after removing redundancies. Recursive feature elimination resulted in the following top 4 discriminative features: gray-level size zone matrix-derived size zone non-uniformity from pre-contrast and delayed phases, gray-level dependency matrix-derived large dependence high gray-level emphasis from venous-phase, and gray-level co-occurrence matrix-derived cluster shade from delayed-phase. A binary classification model with leave-one-out cross-validation showed AUC = 0.85, sensitivity = 84.2%, and specificity = 71.4%. Machine learning and radiomic features extraction can differentiate between benign and malignant indeterminate adrenal tumors and can be used to direct further workup with high sensitivity and specificity.

Volatile-Nonvolatile Memory Network for Progressive Image Super-Resolution

Single-image super-resolution, i.e., reconstructing a high-resolution image from a low-resolution image, is a critical concern in many computer vision applications. Recent deep learning-based image super-resolution methods employ massive numbers of model parameters to obtain quality gain. However, this leads to increased model size and high computational complexity. To mitigate this, some methods employ recursive parameter-sharing for better parameter efficiency. Nevertheless, their designs do not adequately exploit the potential of the recursive operation. In this paper, we propose a novel super-resolution method, called a volatile-nonvolatile memory network (VMNet), to maximize the usefulness of the recursive architecture. Specifically, we design two central components called volatile and nonvolatile memories. By means of these, the recursive feature extraction portion of our model performs effective recursive operations that gradually enhance image quality. Through extensive experiments on $\times 2$ , $\times 3$ , and $\times 4$ super-resolution tasks, we demonstrate that our method outperforms existing state-of-the-art methods in terms of image quality and complexity via stable progressive super-resolution.

The application of R language in the selection of characteristic bands for the prediction of protein content in milk powder by Near Infrared Spectroscopy

Milk powder is an important food, suitable for preservation and transportation. Protein is an important nutritional component in milk powder. At present, the routine physical and chemical analysis method is usually used in the detection of components of milk powder, which is time-consuming and labor-consuming. Therefore, it is very important to carry out the rapid non-destructive detection of milk powder quality. The application of near infrared spectroscopy (NIRS) technology in the rapid detection of milk powder quality is increasingly mature, but many analytical techniques are not perfect. In this study, the models of prediction protein in milk powder were established by R language with NIRS, and the characteristic bands were selected by recursive feature extraction (RFE) in R language, and the selected bands were screened one by one to determine their importance. Finally, using Partial least squares (PLS), generalize linear model (GLM), support vector machine (SVM), least angle regression (LARS), linear model (LM) and other methods to build the prediction models with 8 characteristic bands, the R-squared of the models can be increased by 0.1 to 0.7, the model robustness greatly improved.

A compound attributes-based predictive model for drug induced liver injury in humans.

Drug induced liver injury (DILI) is one of the key safety concerns in drug development. To assess the likelihood of drug candidates with potential adverse reactions of liver, we propose a compound attributes-based approach to predicting hepatobiliary disorders that are routinely reported to US Food and Drug Administration (FDA) Adverse Event Reporting System (FAERS). Specifically, we developed a support vector machine (SVM) model with recursive feature extraction, based on physicochemical and structural properties of compounds as model input. Cross validation demonstrates that the predictive model has a robust performance with averaged 70% of both sensitivity and specificity over 500 trials. An independent validation was performed on public benchmark drugs and the results suggest potential utility of our model for identifying safety alerts. This in silico approach, upon further validation, would ultimately be implemented, together with other in vitro safety assays, for screening compounds early in drug development.

A Very Deep Densely Connected Network for Compressed Sensing MRI

Convolutional neural network (CNN) has achieved great success in the compressed sensing-based magnetic resonance imaging (CS-MRI). Latest deep networks for CS-MRI usually consist of a stack of sub-networks, each of which refines the former image prediction to a more accurate one. However, as the sub-network number increases, the information in prior sub-networks has a little influence on subsequent ones, which increases the training difficulties and limits the reconstruction performance of the deep model. In this paper, we propose a novel network, named very deep densely connected network (VDDCN), for CS-MRI. Dense connections are introduced to connect any two sub-networks of VDDCN, so each sub-network can make full use of all former predictions, boosting the reconstruction performance of the whole network. The sub-network of VDDCN is composed of feature extraction and fusion block (FEFB) processing data in the image domain and data consistency (DC) layer enforcing the data fidelity in k-space. Specifically, in FEFB, multi-level features are extracted by the recursive feature extraction and fusion sub-blocks (RFEFSBs) and fused locally to obtain the compact features. The VDDCN is much deeper than the prior deep learning models and able to discover more MR image details. The experimental results demonstrate that our proposed VDDCN outperforms other state-of-the-art CS-MRI methods.

Estimation of stadium construction schedule based on big data analysis

ABSTRACTThere are many constraints on the progress of the construction of stadiums, which leads to the difficulty of schedule estimation. Aiming at this problem, a method of estimating the progress of stadium construction based on big data analysis is proposed. Firstly, the constrained parametric index analysis model of the progress analysis is constructed, and the Gaussian–Markov model of the progress constraint is established. The phase-space reconstruction of the constraint index information data of the construction progress is carried out. Then, an improved quantitative recursive feature extraction algorithm is used to realize the information entropy feature extraction of constraint feature information. Combined with big data information fusion and K-means clustering algorithm, the parameter cluster analysis of the stadiums construction progress parameters is achieved to improve the accuracy of prediction. On this basis, the corresponding manpower, scheduling, resource demand planning and construction preparation plan are compiled, to achieve the purpose of progress control, and effective arrange the progress of construction. Finally, the simulation results show that the prediction accuracy of the stadium construction is high, which is superior in convergence and efficiency, and has good engineering significance.

Automatic Recognition of Long Period Events From Volcano Tectonic Earthquakes at Cotopaxi Volcano

Geophysics experts are interested in understanding the behavior of volcanoes and forecasting possible eruptions by monitoring and detecting the increment on volcano-seismic activity, with the aim of safeguarding human lives and material losses. This paper presents an automatic volcanic event detection and classification system, which considers feature extraction and feature selection stages, to reduce the processing time toward a reliable real-time volcano early warning system (RT-VEWS). We built the proposed approach in terms of the seismicity presented in 2009 and 2010 at the Cotopaxi Volcano located in Ecuador. In the detection stage, the recordings were time segmented by using a nonoverlapping 15-s window, and in the classification stage, the detected seismic signals were 1-min long. For each detected signal conveying seismic events, a comprehensive set of statistical, temporal, spectral, and scale-domain features were compiled and extracted, aiming to separate long-period (LP) events from volcano-tectonic (VT) earthquakes. We benchmarked two commonly used types of feature selection techniques, namely, wrapper (recursive feature extraction) and embedded (cross-validation and pruning). Each technique was used within a suitable and appropriate classification algorithm, either the support vector machine (SVM) or the decision trees. The best result was obtained by using the SVM classifier, yielding up to 99% accuracy in the detection stage and 97% accuracy and sensitivity in the event classification stage. Selected features and their interpretation were consistent among different input spaces in simple terms of the spectral content of the frequency bands at 3.1 and 6.8 Hz. A comparative analysis showed that the most relevant features for automatic discrimination between LP and VT events were one in the time domain, five in the frequency domain, and nine in the scale domain. Our study provides the framework for an event classification system with high accuracy and reduced computational requirements, according to the orientation toward a future RT-VEWS.