Improving Prediction of Chronic Kidney Disease Using KNN Imputed SMOTE Features and TrioNet Model

Chronic kidney disease(CKD)is a major health concern today,requiring early and accurate diagnosis.Machine learning has emerged as a powerful tool for disease detection,and medical professionals are increasingly using ML classifier algorithms to identify CKD early.This study explores the application of advanced machine learning techniques on a CKD dataset obtained from the University of California,UC Irvine Machine Learning repository.The research introduces TrioNet,an ensemble model combining extreme gradient boosting,random forest,and extra tree classifier,which excels in providing highly accurate predictions for CKD.Furthermore,K nearest neighbor(KNN)imputer is utilized to deal withmissing values while synthetic minority oversampling(SMOTE)is used for class-imbalance problems.To ascertain the efficacy of the proposed model,a comprehensive comparative analysis is conducted with various machine learning models.The proposed TrioNet using KNN imputer and SMOTE outperformed other models with 98.97%accuracy for detectingCKD.This in-depth analysis demonstrates the model’s capabilities and underscores its potential as a valuable tool in the diagnosis of CKD.

Depression Intensity Classification from Tweets Using Fast Text Based Weighted Soft Voting Ensemble

Predicting depression intensity from microblogs and social media posts has numerous benefits and applications,including predicting early psychological disorders and stress in individuals or the general public.A major challenge in predicting depression using social media posts is that the existing studies do not focus on predicting the intensity of depression in social media texts but rather only perform the binary classification of depression and moreover noisy data makes it difficult to predict the true depression in the social media text.This study intends to begin by collecting relevant Tweets and generating a corpus of 210000 public tweets using Twitter public application programming interfaces(APIs).A strategy is devised to filter out only depression-related tweets by creating a list of relevant hashtags to reduce noise in the corpus.Furthermore,an algorithm is developed to annotate the data into three depression classes:‘Mild,’‘Moderate,’and‘Severe,’based on International Classification of Diseases-10(ICD-10)depression diagnostic criteria.Different baseline classifiers are applied to the annotated dataset to get a preliminary idea of classification performance on the corpus.Further FastText-based model is applied and fine-tuned with different preprocessing techniques and hyperparameter tuning to produce the tuned model,which significantly increases the depression classification performance to an 84%F1 score and 90%accuracy compared to baselines.Finally,a FastText-based weighted soft voting ensemble(WSVE)is proposed to boost the model’s performance by combining several other classifiers and assigning weights to individual models according to their individual performances.The proposed WSVE outperformed all baselines as well as FastText alone,with an F1 of 89%,5%higher than FastText alone,and an accuracy of 93%,3%higher than FastText alone.The proposed model better captures the contextual features of the relatively small sample class and aids in the detection of early depression intensity prediction from tweets with impactful performances.

Gastrointestinal Diseases Classification Using Deep Transfer Learning and Features Optimization

Gastrointestinal diseases like ulcers, polyps’, and bleeding areincreasing rapidly in the world over the last decade. On average 0.7 millioncases are reported worldwide every year. The main cause of gastrointestinaldiseases is a Helicobacter Pylori (H. Pylori) bacterium that presents in morethan 50% of people around the globe. Many researchers have proposeddifferent methods for gastrointestinal disease using computer vision techniques.Few of them focused on the detection process and the rest of themperformed classification. The major challenges that they faced are the similarityof infected and healthy regions that misleads the correct classificationaccuracy. In this work, we proposed a technique based on Mask Recurrent-Convolutional Neural Network (R-CNN) and fine-tuned pre-trainedResNet-50 and ResNet-152 networks for feature extraction. Initially, the region ofinterest is detected using Mask R-CNN which is later utilized for the trainingof fine-tuned models through transfer learning. Features are extracted fromfine-tuned models that are later fused using a serial approach. Moreover, anImproved Ant Colony Optimization (ACO) algorithm has also opted for thebest feature selection from the fused feature vector. The best-selected featuresare finally classified using machine learning techniques. The experimentalprocess was conducted on the publicly available dataset and obtained animproved accuracy of 96.43%. In comparison with state-of-the-art techniques,it is observed that the proposed accuracy is improved.

Smart Techniques for LULC Micro Class Classification Using Landsat8 Imagery

Wheat species play important role in the price of products and wheat production estimation.There are several mathematical models used for the estimation of the wheat crop but these models are implemented without considering the wheat species which is an important independent variable.The task of wheat species identification is challenging both for human experts as well as for computer vision-based solutions.With the use of satellite remote sensing,it is possible to identify and monitor wheat species on a large scale at any stage of the crop life cycle.In this work,nine popular wheat species are identified by using Landsat8 operational land imager(OLI)and thermal infrared sensor(TIRS)data.Two thousand samples of eachwheat crop species are acquired every fifteen days with a temporal resolution of ten multispectral bands(band two to band eleven).This study employs random forest(RF),artificial neural network,support vector machine,Naive Bayes,and logistic regression for nine types of wheat classification.In addition,deep neural networks are also developed.Experimental results indicate that RF shows the best performance of 91%accuracy while DNN obtains a 90.2%accuracy.Results suggest that remotely sensed data can be used in wheat type estimation and to improve the performance of the mathematical models.

Deep Learning Approach for Automatic Cardiovascular Disease Prediction Employing ECG Signals

Cardiovascular problems have become the predominant cause of death worldwide and a rise in the number of patients has been observed lately.Currently,electrocardiogram(ECG)data is analyzed by medical experts to determine the cardiac abnormality,which is time-consuming.In addition,the diagnosis requires experienced medical experts and is error-prone.However,automated identification of cardiovascular disease using ECGs is a challenging problem and state-of-the-art performance has been attained by complex deep learning architectures.This study proposes a simple multilayer perceptron(MLP)model for heart disease prediction to reduce computational complexity.ECG dataset containing averaged signals with window size 10 is used as an input.Several competing deep learning and machine learning models are used for comparison.K-fold cross-validation is used to validate the results.Experimental outcomes reveal that the MLP-based architecture can produce better outcomes than existing approaches with a 94.40%accuracy score.The findings of this study show that the proposed system achieves high performance indicating that it has the potential for deployment in a real-world,practical medical environment.

Central Aggregator Intrusion Detection System for Denial of Service Attacks

Vehicle-to-grid technology is an emerging field that allows unused power from Electric Vehicles(EVs)to be used by the smart grid through the central aggregator.Since the central aggregator is connected to the smart grid through a wireless network,it is prone to cyber-attacks that can be detected and mitigated using an intrusion detection system.However,existing intrusion detection systems cannot be used in the vehicle-to-grid network because of the special requirements and characteristics of the vehicle-to-grid network.In this paper,the effect of denial-of-service attacks of malicious electric vehicles on the central aggregator of the vehicle-to-grid network is investigated and an intrusion detection system for the vehicle-to-grid network is proposed.The proposed system,central aggregator–intrusion detection system(CA-IDS),works as a security gateway for EVs to analyze andmonitor incoming traffic for possible DoS attacks.EVs are registered with a Central Aggregator(CAG)to exchange authenticated messages,and malicious EVs are added to a blacklist for violating a set of predefined policies to limit their interaction with the CAG.A denial of service(DoS)attack is simulated at CAG in a vehicle-to-grid(V2G)network manipulating various network parameters such as transmission overhead,receiving capacity of destination,average packet size,and channel availability.The proposed system is compared with existing intrusion detection systems using different parameters such as throughput,jitter,and accuracy.The analysis shows that the proposed system has a higher throughput,lower jitter,and higher accuracy as compared to the existing schemes.

Drug Usage Safety from Drug Reviews with Hybrid Machine Learning Approach

With the increasing usage of drugs to remedy different diseases,drug safety has become crucial over the past few years.Often medicine from several companies is offered for a single disease that involves the same/similar substances with slightly different formulae.Such diversification is both helpful and danger-ous as such medicine proves to be more effective or shows side effects to different patients.Despite clinical trials,side effects are reported when the medicine is used by the mass public,of which several such experiences are shared on social media platforms.A system capable of analyzing such reviews could be very helpful to assist healthcare professionals and companies for evaluating the safety of drugs after it has been marketed.Sentiment analysis of drug reviews has a large poten-tial for providing valuable insights into these cases.Therefore,this study proposes an approach to perform analysis on the drug safety reviews using lexicon-based and deep learning techniques.A dataset acquired from the‘Drugs.Com’contain-ing reviews of drug-related side effects and reactions,is used for experiments.A lexicon-based approach,Textblob is used to extract the positive,negative or neu-tral sentiment from the review text.Review classification is achieved using a novel hybrid deep learning model of convolutional neural networks and long short-term memory(CNN-LSTM)network.The CNN is used at thefirst level to extract the appropriate features while LSTM is used at the second level.Several well-known machine learning models including logistic regression,random for-est,decision tree,and AdaBoost are evaluated using term frequency-inverse docu-ment frequency(TF-IDF),a bag of words(BoW),feature union of(TF-IDF+BoW),and lexicon-based methods.Performance analysis with machine learning models,long short term memory and convolutional neural network models,and state-of-the-art approaches indicate that the proposed CNN-LSTM model shows superior performance with an 0.96 accuracy.We also performed a statistical sig-nificance T-test to show the significance of the proposed CNN-LSTM model in comparison with other approaches.

Chi-Square and PCA Based Feature Selection for Diabetes Detection with Ensemble Classifier

Diabetes mellitus is a metabolic disease that is ranked among the top 10 causes of death by the world health organization.During the last few years,an alarming increase is observed worldwide with a 70%rise in the disease since 2000 and an 80%rise in male deaths.If untreated,it results in complications of many vital organs of the human body which may lead to fatality.Early detection of diabetes is a task of significant importance to start timely treatment.This study introduces a methodology for the classification of diabetic and normal people using an ensemble machine learning model and feature fusion of Chi-square and principal component analysis.An ensemble model,logistic tree classifier(LTC),is proposed which incorporates logistic regression and extra tree classifier through a soft voting mechanism.Experiments are also performed using several well-known machine learning algorithms to analyze their performance including logistic regression,extra tree classifier,AdaBoost,Gaussian naive Bayes,decision tree,random forest,and k nearest neighbor.In addition,several experiments are carried out using principal component analysis(PCA)and Chi-square(Chi-2)fea-tures to analyze the influence of feature selection on the performance of machine learning classifiers.Results indicate that Chi-2 features show high performance than both PCA features and original features.However,the highest accuracy is obtained when the proposed ensemble model LTC is used with the proposed fea-ture fusion framework-work which achieves a 0.85 accuracy score which is the highest of the available approaches for diabetes prediction.In addition,the statis-tical T-test proves the statistical significance of the proposed approach over other approaches.

Fast Intra Mode Selection in HEVC Using Statistical Model

Comprehension algorithms like High Efficiency Video Coding(HEVC)facilitates fast and efficient handling of multimedia contents.Such algorithms involve various computation modules that help to reduce the size of content but preserve the same subjective viewing quality.However,the brute-force behavior of HEVC is the biggest hurdle in the communication of multimedia content.Therefore,a novel method will be presented here to accelerate the encoding process of HEVC by making early intra mode decisions for the block.Normally,the HEVC applies 35 intra modes to every block of the frame and selects the best among them based on the RD-cost(rate-distortion).Firstly,the proposed work utilizes neighboring blocks to extract available information for the current block.Then this information is converted to the probability that tells which intra mode might be best in the current situation.The proposed model has a strong foundation as it is based on the probability rule-2 which says that the sum of probabilities of all outcomes should be 1.Moreover,it is also based on optimal stopping theory(OST).Therefore,the proposed model performs better than many existing OST and classical secretary-basedmodels.The proposed algorithms expedited the encoding process by 30.22%of the HEVC with 1.35%Bjontegaard Delta Bit Rate(BD-BR).

Ensembling Neural Networks for User’s Indoor Localization Using Magnetic Field Data from Smartphones

Predominantly the localization accuracy of the magnetic field-based localization approaches is severed by two limiting factors:Smartphone heterogeneity and smaller data lengths.The use of multifarioussmartphones cripples the performance of such approaches owing to the variability of the magnetic field data.In the same vein,smaller lengths of magnetic field data decrease the localization accuracy substantially.The current study proposes the use of multiple neural networks like deep neural network(DNN),long short term memory network(LSTM),and gated recurrent unit network(GRN)to perform indoor localization based on the embedded magnetic sensor of the smartphone.A voting scheme is introduced that takes predictions from neural networks into consideration to estimate the current location of the user.Contrary to conventional magnetic field-based localization approaches that rely on the magnetic field data intensity,this study utilizes the normalized magnetic field data for this purpose.Training of neural networks is carried out using Galaxy S8 data while the testing is performed with three devices,i.e.,LG G7,Galaxy S8,and LG Q6.Experiments are performed during different times of the day to analyze the impact of time variability.Results indicate that the proposed approach minimizes the impact of smartphone variability and elevates the localization accuracy.Performance comparison with three approaches reveals that the proposed approach outperforms them in mean,50%,and 75%error even using a lesser amount of magnetic field data than those of other approaches.

A Multilevel Deep Feature Selection Framework for Diabetic Retinopathy Image Classification

Diabetes or Diabetes Mellitus(DM)is the upset that happens due to high glucose level within the body.With the passage of time,this polygenic disease creates eye deficiency referred to as Diabetic Retinopathy(DR)which can cause a major loss of vision.The symptoms typically originate within the retinal space square in the form of enlarged veins,liquid dribble,exudates,haemorrhages and small scale aneurysms.In current therapeutic science,pictures are the key device for an exact finding of patients’illness.Meanwhile,an assessment of new medicinal symbolisms stays complex.Recently,Computer Vision(CV)with deep neural networks can train models with high accuracy.The thought behind this paper is to propose a computerized learning model to distinguish the key precursors of Dimensionality Reduction(DR).The proposed deep learning framework utilizes the strength of selected models(VGG and Inception V3)by fusing the extracated features.To select the most discriminant features from a pool of features,an entropy concept is employed before the classification step.The deep learning models are fit for measuring the highlights as veins,liquid dribble,exudates,haemorrhages and miniaturized scale aneurysms into various classes.The model will ascertain the loads,which give the seriousness level of the patient’s eye.The model will be useful to distinguish the correct class of seriousness of diabetic retinopathy pictures.

Prediction Models for COVID-19 Integrating Age Groups, Gender, and Underlying Conditions

The COVID-19 pandemic has caused hundreds of thousands of deaths,millions of infections worldwide,and the loss of trillions of dollars for many large economies.It poses a grave threat to the human population with an excessive number of patients constituting an unprecedented challenge with which health systems have to cope.Researchers from many domains have devised diverse approaches for the timely diagnosis of COVID-19 to facilitate medical responses.In the same vein,a wide variety of research studies have investigated underlying medical conditions for indicators suggesting the severity and mortality of,and role of age groups and gender on,the probability of COVID-19 infection.This study aimed to review,analyze,and critically appraise published works that report on various factors to explain their relationship with COVID-19.Such studies span a wide range,including descriptive analyses,ratio analyses,cohort,prospective and retrospective studies.Various studies that describe indicators to determine the probability of infection among the general population,as well as the risk factors associated with severe illness and mortality,are critically analyzed and these ndings are discussed in detail.A comprehensive analysis was conducted on research studies that investigated the perceived differences in vulnerability of different age groups and genders to severe outcomes of COVID-19.Studies incorporating important demographic,health,and socioeconomic characteristics are highlighted to emphasize their importance.Predominantly,the lack of an appropriated dataset that contains demographic,personal health,and socioeconomic information implicates the efcacy and efciency of the discussed methods.Results are overstated on the part of both exclusion of quarantined and patients with mild symptoms and inclusion of the data from hospitals where the majority of the cases are potentially ill.

MagneFi: Multiuser, Multi-Building and Multi-Floor Geomagnetic Field Dataset for Indoor Positioning

Indoor positioning and localization have emerged as a potential research area during the last few years owing to the wide proliferation of smartphones and the inception of location-attached services for the consumer industry.Due to the importance of precise location information,several positioning technologies are adopted such as Wi-Fi,ultrawideband,infrared,radio frequency identification,Bluetooth beacons,pedestrian dead reckoning,and magnetic field,etc.Although Wi-Fi and magnetic field-based positioning are more attractive concerning the deployment of Wi-Fi access points and ubiquity of magnetic field data,the latter is preferred as it does not require any additional infrastructure as other approaches do.Despite the advantages of magnetic field positioning,comparing the performance of positioning and localization algorithms is very difficult due to the lack of good public datasets that cover various aspects of the magnetic field data.Available datasets do not provide the data to analyze the impact of device heterogeneity,user heights,and time-specific magnetic field mutation.Moreover,multi-floor and multibuilding data are available for the evaluation of state-of-the-art approaches.To overcome the above-mentioned issues,this study presents multi-user,multidevice,multi-building magnetic field data which is collected over a longer period.The dataset contains the data from five different smartphones including Samsung Galaxy S8,S9,A8,LG G6,and LG G7 for three geographically separated buildings.Three users including one female and two males collected the data for various path geometry and data collection scenarios.Moreover,the data contains the magnetic field samples collected on stairs to test multifloor localization.Besides the magnetic field data,the data from inertial measurement unit sensors like the accelerometer,motion sensors,and barometer is provided as well.

Wi-Fi Positioning Dataset with Multiusers and Multidevices Considering Spatio-Temporal Variations

Precise information on indoor positioning provides a foundation for position-related customer services.Despite the emergence of several indoor positioning technologies such as ultrawideband,infrared,radio frequency identification,Bluetooth beacons,pedestrian dead reckoning,and magnetic field,Wi-Fi is one of the most widely used technologies.Predominantly,Wi-Fi fingerprinting is the most popular method and has been researched over the past two decades.Wi-Fi positioning faces three core problems:device heterogeneity,robustness to signal changes caused by human mobility,and device attitude,i.e.,varying orientations.The existing methods do not cover these aspects owing to the unavailability of publicly available datasets.This study introduces a dataset that includes the Wi-Fi received signal strength(RSS)gathered using four different devices,namely Samsung Galaxy S8,S9,A8,LG G6,and LG G7,operated by three surveyors,including a female and two males.In addition,three orientations of the smartphones are used for the data collection and include multiple buildings with a multifloor environment.Various levels of human mobility have been considered in dynamic environments.To analyze the time-related impact on Wi-Fi RSS,data over 3 years have been considered.

Malicious Traffic Detection in IoT and Local Networks Using Stacked Ensemble Classifier

Malicious traffic detection over the internet is one of the challenging areas for researchers to protect network infrastructures from any malicious activity.Several shortcomings of a network system can be leveraged by an attacker to get unauthorized access through malicious traffic.Safeguard from such attacks requires an efficient automatic system that can detect malicious traffic timely and avoid system damage.Currently,many automated systems can detect malicious activity,however,the efficacy and accuracy need further improvement to detect malicious traffic from multi-domain systems.The present study focuses on the detection of malicious traffic with high accuracy using machine learning techniques.The proposed approach used two datasets UNSW-NB15 and IoTID20 which contain the data for IoT-based traffic and local network traffic,respectively.Both datasets were combined to increase the capability of the proposed approach in detecting malicious traffic from local and IoT networks,with high accuracy.Horizontally merging both datasets requires an equal number of features which was achieved by reducing feature count to 30 for each dataset by leveraging principal component analysis(PCA).The proposed model incorporates stacked ensemble model extra boosting forest(EBF)which is a combination of tree-based models such as extra tree classifier,gradient boosting classifier,and random forest using a stacked ensemble approach.Empirical results show that EBF performed significantly better and achieved the highest accuracy score of 0.985 and 0.984 on the multi-domain dataset for two and four classes,respectively.

Hyoscine for polyp detection during colonoscopy: A meta-analysis and systematic review

AIM: To assess the role of hyoscine for polyp detectionduring colonoscopy.METHODS: Studies(randomized controlled trials orRCTs) that compared the use of hyoscine vs no hyo-scine or placebo for polyp detection during colonoscopywere included in our analysis. A search on multiple da-tabases was performed in September 2013 with searchterms being "hyoscine and colonoscopy", "hyoscineand polyp", "hyoscine and adenoma", "antispasmoticand colonoscopy", "antispasmotic and adenoma", and"antispasmotic and polyp". Jadad scoring was used toassess the quality of studies. The efficacy of hyoscinewas analyzed using Mantel-Haenszel model for polypand adenoma detection with odds ratio(OR). The I2measure of inconsistency was used to assess hetero-geneity(P < 0.05 or I2 > 50%). Statistical analysis was performed by RevMan 5.1. Funnel plots was used to assess publication bias.RESULTS: The search of the electronic databases identified 283 articles. Of these articles, eight published RCTs performed at various locations in Europe, Asia, and Australia were included in our meta-analysis, seven published as manuscripts and one published as an ab-stract(n = 2307). All the studies included patients with a hyoscine and a no hyoscine/placebo group and were of adequate quality(Jadad score ≥ 2). Eight RCTs as-sessed the polyp detection rate(PDR)(n = 2307). The use of hyoscine demonstrated no statistically significant difference as compared to no hyoscine or placebo for PDR(OR = 1.06; 95%CI: 0.89-1.25; P = 0.51). Five RCTs assessed the adenoma detection rate(ADR)(n = 2015). The use of hyoscine demonstrated no statisti-cally significant difference as compared to no hyoscine or placebo for ADR(OR = 1.12; 95%CI: 0.92-1.37; P = 0.25). Furthermore, the timing of hyoscine admin-istration(given at cecal intubation or pre-procedure) demonstrated no differences in PDR compared to no hyoscine or placebo. Publication bias or heterogeneity was not observed for any of the outcomes.CONCLUSION: Hyoscine use in patients undergoing colonoscopy does not appear to significantly increase the detection of polyps or adenomas.

Ascorbic acid and low-volume polyethylene glycol for bowel preparation prior to colonoscopy:A meta-analysis

AIM: To evaluate the benefits of low-volume polyethylene glycol(PEG) with ascorbic acid compared to fulldose PEG for colonoscopy preparation. METHODS: MEDLINE, Cochrane Central Register of Controlled Trials and Database of Systematic Reviews, CINAHL, Pub Med, and recent abstracts from major conferences were searched(January 2012). Only randomized-controlled trials on adult subjects comparing lowvolume PEG(2 L) with ascorbic acid vs full-dose PEG(3 or 4 L) were included. Meta-analysis for the efficacy of low-volume PEG with ascorbic acid and full-dose PEG were analyzed by calculating pooled estimates of number of satisfactory bowel preparations as well as adverse patient events(abdominal pain, nausea, vomiting). Separate analyses were performed for each main outcome by using OR with fixed and random effects models. Heterogeneity was assessed by calculating the I2 measure of inconsistency. Rev Man 5.1 was utilized for statistical analysis.RESULTS: The initial search identified 242 articles and trials. Nine studies(n = 2911) met the inclusion criteria and were analyzed for this meta-analysis with mean age range from 53.0 to 59.6 years. All studies were randomized controlled trials on adult patients comparing large-volume PEG solutions(3 or 4 L) with low-volume PEG solutions and ascorbic acid. No statistically significant difference was noted between lowvolume PEG with ascorbic acid and full-dose PEG for number of satisfactory bowel preparations(OR 1.07, 95%CI: 0.86-1.33, P = 0.56). No statistically significant difference was noted between low-volume PEG with ascorbic acid and full-dose PEG for abdominal pain(OR 1.09, 95%CI: 0.81-1.48, P = 0.56), nausea(OR 0.70, 95%CI: 0.49-1.00, P = 0.05), or vomiting(OR 0.99, 95%CI: 0.78-1.26, P = 0.95). No publication bias was noted.CONCLUSION: Low-volume PEG with the addition of ascorbic acid demonstrates no statistically significant difference to full-dose PEG for satisfactory bowel preparation and side-effects.