An Ensemble approach for feature selection and classification in intrusion detection using Extra-Tree algorithm

The number of attacks increased with speedy development in web communication in the last couple of years. The Anomaly Detection method for IDS has become substantial in detecting novel attacks in Intrusion Detection System (IDS). Achieving high accuracy are the significant challenges in designing an intrusion detection system. It also emphasizes applying different feature selection techniques to identify the most suitable feature subset. The author uses Extremely randomized trees (Extra-Tree) for feature importance. The author tries multiple thresholds on the feature importance parameters to find the best features. If single classifiers use, then the classifier's output is wrong, so that the final decision may be wrong. So The author uses an Extra-Tree classifier applied to the best-selected features. The proposed method is estimated on standard datasets KDD CUP'99, NSL-KDD, and UNSW-NB15. The experimental results show that the proposed approach performs better than existing methods in detection rate, false alarm rate, and accuracy.

Land Cover Dynamics on the Lower Ganges–Brahmaputra Delta: Agriculture–Aquaculture Transitions, 1972–2017

Aquaculture in tropical and subtropical developing countries has expanded in recent years. This practice is controversial due to its potential for serious economic, food security, and environmental impacts—especially for intensive operations in and near mangrove ecosystems, where many shrimp species spawn. While considerable effort has been directed toward understanding aquaculture impacts, maps of spatial extent and multi-decade spatiotemporal dynamics remain sparse. This is in part because aquaculture ponds (ghers) can be challenging to distinguish from other shallow water targets on the basis of water-leaving radiance alone. Here, we focus on the Lower Ganges–Brahmaputra Delta (GBD), one of the most expansive areas of recent aquaculture growth on Earth and adjacent to the Sundarbans mangrove forest, a biodiversity hotspot. We use a combination of MODIS 16-day EVI composites and 45 years (1972–2017) of Landsat observations to characterize dominant spatiotemporal patterns in the vegetation phenology of the area, identify consistent seasonal optical differences between flooded ghers and other land uses, and quantify the multi-decade expansion of standing water bodies. Considerable non-uniqueness exists in the spectral signature of ghers on the GBD, propagating into uncertainty in estimates of spatial extent. We implement three progressive decision boundaries to explicitly quantify this uncertainty and provide liberal, moderate, and conservative estimates of flooded gher extent on three different spatial scales. Using multiple extents and multiple thresholds, we quantify the size distribution of contiguous regions of flooded gher extent at ten-year intervals. The moderate threshold shows standing water area within Bangladeshi polders to have expanded from less than 300 km2 in 1990 to over 1400 km2 in 2015. At all three scales investigated, the size distribution of standing water bodies is increasingly dominated by larger, more interconnected networks of flooded areas associated with aquaculture. Much of this expansion has occurred in immediate proximity to the Bangladeshi Sundarbans.

Serum Neuron-Specific Enolase Thresholds for Predicting Postcardiac Arrest Outcome

Background and ObjectivesTo determine thresholds of serum neuron-specific enolase (NSE) for prediction of poor outcome after cardiac arrest with >95% specificity using a unique method of multiple thresholds meta-analysis.MethodsData from a systematic review by the European Resuscitation Council (ERC 2014) were updated with literature searches from PubMed, Cochrane, and Scopus until August 2020. Search terms included the MeSH terms “heart arrest” and “biomarkers” and the text words “cardiac arrest,” “neuron specific enolase,” “coma” and “prognosis.” Cohort studies with comatose cardiac arrest survivors aged >16 years undergoing targeted temperature management (TTM) and NSE levels within 96 hours of resuscitation were included. Poor outcome was defined as cerebral performance category 3–5 at hospital discharge or later. Studies without extractable contingency tables were excluded. A multiple thresholds meta-analysis model was used to generate summary receiver operating characteristic curves for various time points. NSE thresholds (and 95% prediction intervals) for >95% specificity were calculated. Evidence appraisal was performed using a method adapted from the American Academy of Neurology grading criteria.ResultsData from 11 studies (n = 1,982) at 0–24 hours, 21 studies (n = 2,815) at 24–48 hours, and 13 studies (n = 2,557) at 48–72 hours was analyzed. Areas under the curve for prediction of poor outcomes were significantly larger at 24–48 hours and 48–72 hours compared to 0–24 hours (0.82 and 0.83 vs 0.64). Quality of evidence was very low for most studies because of the risk of incorporation bias—knowledge of NSE levels potentially influenced life support withdrawal decisions. To minimize falsely pessimistic predictions, NSE thresholds at the upper 95% limit of prediction intervals are reported. For prediction of poor outcome with specificity >95%, upper limits of the prediction interval for NSE were 70.4 ng/mL at 24–48 hours and 58.6 ng/mL at 48–72 hours. Sensitivity analyses excluding studies with inconsistent TTM use or different outcome criteria did not substantially alter the results.ConclusionsNSE thresholds for highly specific prediction of poor outcome are much higher than generally used. Future studies must minimize bias by masking treatment teams to the results of potential predictors and by prespecifying criteria for withdrawal of life support.

COMPARISON OF PEROMETRY-BASED VOLUMETRIC ARM MEASUREMENTS AND BIOIMPEDANCE SPECTROSCOPY FOR EARLY IDENTIFICATION OF LYMPHEDEMA IN A PROSPECTIVELY-SCREENED COHORT OF BREAST CANCER PATIENTS

Breast cancer-related lymphedema (BCRL) affects more than one in five women treated for breast cancer, and women remain at lifelong risk. Screening for BCRL is recommended by several national and international organizations for women at risk of BCRL, and multiple methods of objective screening measurement exist. The goal of this study was to compare the use of perometry and bioimpedance spectroscopy (BIS) for early identification of BCRL in a cohort of 138 prospectively screened patients. At each screening visit, a patient's relative volume change (RVC) from perometer measurements and change in L-Dex from baseline (ΔL-Dex) using BIS was calculated. There was a negligible correlation between RVC and ΔL-Dex (r=0.195). Multiple thresholds of BCRL were examined: RVC ≥5% and ≥10% as well as and ΔL-Dex ≥6.5 and ≥10. While some patients developed an elevated RVC and ΔL-Dex, many demonstrated elevations in only one threshold category. Moreover, the majority of patients with RVC ≥5%, ΔL-Dex ≥6.5, or ΔL-Dex ≥10 regressed to non-elevated measurements without intervention. These findings suggest a role for combining multiple screening methods for early identification of BCRL; furthermore, BCRL diagnosis must incorporate patient symptoms and clinical evaluation with objective measurements obtained from techniques such as perometry and bioimpedance spectroscopy.

Multiple thresholds and trajectories of microbial biodiversity predicted across browning gradients by neural networks and decision tree learning

Ecological association studies often assume monotonicity such as between biodiversity and environmental properties although there is growing evidence that nonmonotonic relations dominate in nature. Here, we apply machine-learning algorithms to reveal the nonmonotonic association between microbial diversity and an anthropogenic-induced large-scale change, the browning of freshwaters, along a longitudinal gradient covering 70 boreal lakes in Scandinavia. Measures of bacterial richness and evenness (alpha-diversity) showed nonmonotonic trends in relation to environmental gradients, peaking at intermediate levels of browning. Depending on the statistical methods, variables indicative for browning could explain 5% of the variance in bacterial community composition (beta-diversity) when applying standard methods assuming monotonic relations and up to 45% with machine-learning methods taking non-monotonicity into account. This non-monotonicity observed at the community level was explained by the complex interchangeable nature of individual taxa responses as shown by a high degree of nonmonotonic responses of individual bacterial sequence variants to browning. Furthermore, the nonmonotonic models provide the position of thresholds and predict alternative bacterial diversity trajectories in boreal freshwater as a result of ongoing climate and land-use changes, which in turn will affect entire ecosystem metabolism and likely greenhouse gas production.

An Improved Cuckoo Search Algorithm for Multithreshold Image Segmentation

Image segmentation is an important part of image processing. For the disadvantages of image segmentation under multiple thresholds such as long time and poor quality, an improved cuckoo search (ICS) is proposed for multithreshold image segmentation strategy. Firstly, the image segmentation model based on the maximum entropy threshold is described, and secondly, the cuckoo algorithm is improved by using chaotic initialization population to improve the diversity of solutions, optimizing the step size factor to improve the possibility of obtaining the optimal solution, and using probability to reduce the complexity of the algorithm; finally, the maximum entropy threshold function in image segmentation is used as the individual fitness function of the cuckoo search algorithm for solving. The simulation experiments show that the algorithm has a good segmentation effect under four different thresholding conditions.

FibroTest for Evaluating Fibrosis in Non-Alcoholic Fatty Liver Disease Patients: A Systematic Review and Meta-Analysis

(1) Background: FibroTest™ is a multi-marker panel, suggested by guidelines as one of the surrogate markers with acceptable performance for detecting fibrosis in patients with non-alcoholic fatty liver disease (NAFLD). A number of studies evaluating this test have been published after publication of the guidelines. This study aims to produce summary estimates of FibroTest™ diagnostic accuracy. (2) Methods: Five databases were searched for studies that evaluated FibroTest™ against liver biopsy as the reference standard in NAFLD patients. Two authors independently screened the references, extracted data, and assessed the quality of included studies. Meta-analyses of the accuracy in detecting different levels of fibrosis were performed using the bivariate random-effects model and the linear mixed-effects multiple thresholds model. (3) Results: From ten included studies, seven were eligible for inclusion in our meta-analysis. Five studies were included in the meta-analysis of FibroTest™ in detecting advanced fibrosis and five in significant fibrosis, resulting in an AUC of 0.77 for both target conditions. The meta-analysis of three studies resulted in an AUC of 0.69 in detecting any fibrosis, while analysis of three other studies showed higher accuracy in cirrhosis (AUC: 0.92). (4) Conclusions: Our meta-analysis showed acceptable performance (AUC > 0.80) of FibroTest™ only in detecting cirrhosis. We observed more limited performance of the test in detecting significant and advanced fibrosis in NAFLD patients. Further primary studies with high methodological quality are required to validate the reliability of the test for detecting different fibrosis levels and to compare the performance of the test in different settings.

Bayesian network models with decision tree analysis for management of childhood malaria in Malawi

Background Malaria is a major cause of death in children under five years old in low- and middle-income countries such as Malawi. Accurate diagnosis and management of malaria can help reduce the global burden of childhood morbidity and mortality. Trained healthcare workers in rural health centers manage malaria with limited supplies of malarial diagnostic tests and drugs for treatment. A clinical decision support system that integrates predictive models to provide an accurate prediction of malaria based on clinical features could aid healthcare workers in the judicious use of testing and treatment. We developed Bayesian network (BN) models to predict the probability of malaria from clinical features and an illustrative decision tree to model the decision to use or not use a malaria rapid diagnostic test (mRDT). Methods We developed two BN models to predict malaria from a dataset of outpatient encounters of children in Malawi. The first BN model was created manually with expert knowledge, and the second model was derived using an automated method. The performance of the BN models was compared to other statistical models on a range of performance metrics at multiple thresholds. We developed a decision tree that integrates predictions with the costs of mRDT and a course of recommended treatment. Results The manually created BN model achieved an area under the ROC curve (AUC) equal to 0.60 which was statistically significantly higher than the other models. At the optimal threshold for classification, the manual BN model had sensitivity and specificity of 0.74 and 0.42 respectively, and the automated BN model had sensitivity and specificity of 0.45 and 0.68 respectively. The balanced accuracy values were similar across all the models. Sensitivity analysis of the decision tree showed that for values of probability of malaria below 0.04 and above 0.40, the preferred decision that minimizes expected costs is not to perform mRDT. Conclusion In resource-constrained settings, judicious use of mRDT is important. Predictive models in combination with decision analysis can provide personalized guidance on when to use mRDT in the management of childhood malaria. BN models can be efficiently derived from data to support clinical decision making.

Virtual partition digital PCR for high precision chromosomal counting applications

Digital PCR (dPCR) is the gold standard analytical platform for rapid high precision quantification of genomic fragments. However, current dPCR assays are generally limited to monitoring 1-2 analytes per sample, thereby limiting the platform’s ability to address some clinical applications that require the simultaneous monitoring of 20 – 50 analytes per sample. Here we present Virtual Partition dPCR (VPdPCR), a novel analysis methodology enabling the detection of 10 or more target regions per color channel using conventional dPCR hardware and workflow. Furthermore, VPdPCR enables dPCR instruments to overcome upper quantitation limits caused by partitioning error. While traditional dPCR analysis establishes a single threshold to separate negative and positive partitions, VPdPCR establishes multiple thresholds to identify the number of unique targets present in each positive droplet based on fluorescent intensity. Each physical partition is then divided into a series of virtual partitions, and the resulting increase in partition count substantially decreases partitioning error. We present both a theoretical analysis of the advantages of VPdPCR and an experimental demonstration in the form of a 20-plex assay for non-invasive fetal aneuploidy testing. This demonstration assay – tested on 432 samples contrived from sheared cell-line DNA at multiple input concentrations and simulated fractions of euploid or trisomy-21 “fetal” DNA – is analyzed using both traditional dPCR thresholding and VPdPCR. VPdPCR analysis significantly lowers variance of chromosome ratio across replicates and increases the accuracy of trisomy identification when compared to traditional dPCR, yielding >98% single-well sensitivity and specificity. VPdPCR has substantial promise for increasing the utility of dPCR in applications requiring ultra-high-precision quantitation.

Neural networks and extreme gradient boosting predict multiple thresholds and trajectories of microbial biodiversity responses due to browning

Ecological association studies often assume monotonicity such as between biodiversity and environmental properties although there is growing evidence that non-monotonic relations dominate in nature. Here we apply machine learning algorithms to reveal the non-monotonic association between microbial diversity and an anthropogenic induced large scale change, the browning of freshwaters, along a longitudinal gradient covering 70 boreal lakes in Scandinavia. Measures of bacterial richness and evenness (alpha diversity) showed non-monotonic trends in relation to environmental gradients, peaking at intermediate levels of browning. Depending on the statistical methods, variables indicative for browning could explain 5% of the variance in bacterial community composition (beta diversity) when applying standard methods assuming monotonic relations and up to 45 % with machine learning methods (i.e. extreme gradient boosting and feed-forward neural networks) taking non-monotonicity into account. This non-monotonicity observed at the community level was explained by the complex interchangeable nature of individual taxa responses as shown by a high degree of non-monotonic responses of individual bacterial sequence variants to browning. Furthermore, the non-monotonic models provide the position of thresholds and predict alternative bacterial diversity trajectories in boreal freshwater as a result of ongoing climate and land use changes, which in turn will affect entire ecosystem metabolism and likely greenhouse gas production.