Privacy Preserved Brain Disorder Diagnosis Using Federated Learning

Federated learning has recently attracted significant attention as a cutting-edge technology that enables Artificial Intelligence(AI)algorithms to utilize global learning across the data of numerous individuals while safeguarding user data privacy.Recent advanced healthcare technologies have enabled the early diagnosis of various cognitive ailments like Parkinson’s.Adequate user data is frequently used to train machine learning models for healthcare systems to track the health status of patients.The healthcare industry faces two significant challenges:security and privacy issues and the personalization of cloud-trained AI models.This paper proposes a Deep Neural Network(DNN)based approach embedded in a federated learning framework to detect and diagnose brain disorders.We extracted the data from the database of Kay Elemetrics voice disordered and divided the data into two windows to create training models for two clients,each with different data.To lessen the over-fitting aspect,every client reviewed the outcomes in three rounds.The proposed model identifies brain disorders without jeopardizing privacy and security.The results reveal that the global model achieves an accuracy of 82.82%for detecting brain disorders while preserving privacy.

Applying Customized Convolutional Neural Network to Kidney Image Volumes for Kidney Disease Detection

Kidney infection is a severe medical issue affecting individuals worldwide and increasing mortality rates.Chronic Kidney Disease(CKD)is treatable during its initial phases but can become irreversible and cause renal failure.Among the various diseases,the most prevalent kidney conditions affecting kidney function are cyst growth,kidney tumors,and nephrolithiasis.The significant challenge for the medical community is the immediate diagnosis and treatment of kidney disease.Kidney failure could result from kidney disorders like tumors,stones,and cysts if not often identified and addressed.Computer-assisted diagnostics are necessary to support clinicians’and specialists’medical assessments due to the rising prevalence of chronic renal illness,the lack of experts,and the rising rates of assessment and monitoring,mainly in developing nations.Artificial Intelligence(AI)approaches such as machine,and deep learning has been used in literature for kidney disease detection;however,they still lack performance.This paper implements a deep learning-based Convolutional Neural Network(CNN)model for the classification and prognosis of kidney disease.We use a benchmark Computed Tomography(CT)kidney dataset for experimentation.The data is pre-processed,and then CNN extracts the features from the images.Results reveal that the proposed approach accurately classifies kidney disease with a considerable accuracy of 0.992%,0.994%precision,0.982%recall,and 0.987%F1-score.This study suggests using the proposed fine-tuned CNN model for kidney disease detection.

Computing of LQR Technique for Nonlinear System Using Local Approximation

The main idea behind the present research is to design a state-feedback controller for an underactuated nonlinear rotary inverted pendulum module by employing the linear quadratic regulator(LQR)technique using local approximation.The LQR is an excellent method for developing a controller for nonlinear systems.It provides optimal feedback to make the closed-loop system robust and stable,rejecting external disturbances.Model-based optimal controller for a nonlinear system such as a rotatory inverted pendulum has not been designed and implemented using Newton-Euler,Lagrange method,and local approximation.Therefore,implementing LQR to an underactuated nonlinear system was vital to design a stable controller.A mathematical model has been developed for the controller design by utilizing the Newton-Euler,Lagrange method.The nonlinear model has been linearized around an equilibrium point.Linear and nonlinear models have been compared to find the range in which linear and nonlinear models’behaviour is similar.MATLAB LQR function and system dynamics have been used to estimate the controller parameters.For the performance evaluation of the designed controller,Simulink has been used.Linear and nonlinear models have been simulated along with the designed controller.Simulations have been performed for the designed controller over the linear and nonlinear system under different conditions through varying system variables.The results show that the system is stable and robust enough to act against external disturbances.The controller maintains the rotary inverted pendulum in an upright position and rejects disruptions like falling under gravitational force or any external disturbance by adjusting the rotation of the horizontal link in both linear and nonlinear environments in a specific range.The controller has been practically designed and implemented.It is vivid from the results that the controller is robust enough to reject the disturbances in milliseconds and keeps the pendulum arm deflection angle to zero degrees.

A Drones Optimal Path Planning Based on Swarm Intelligence Algorithms

The smart city comprises various interlinked elements which communicate data and offers urban life to citizen.Unmanned Aerial Vehicles(UAV)or drones were commonly employed in different application areas like agriculture,logistics,and surveillance.For improving the drone flying safety and quality of services,a significant solution is for designing the Internet of Drones(IoD)where the drones are utilized to gather data and people communicate to the drones of a specific flying region using the mobile devices is for constructing the Internet-of-Drones,where the drones were utilized for collecting the data,and communicate with others.In addition,the SIRSS-CIoD technique derives a tuna swarm algorithm-based clustering(TSA-C)technique to choose cluster heads(CHs)and organize clusters in IoV networks.Besides,the SIRSS-CIoD technique involves the design of a biogeography-based optimization(BBO)technique to an optimum route selection(RS)process.The design of clustering and routing techniques for IoD networks in smart cities shows the novelty of the study.A wide range of experimental analyses is carried out and the comparative study highlighted the improved performance of the SIRSS-CIoD technique over the other approaches.

A Blockchain-Based Architecture for Enabling Cybersecurity in the Internet-of-Critical Infrastructures

Due to the drastic increase in the number of critical infrastructures like nuclear plants,industrial control systems(ICS),transportation,it becomes highly vulnerable to several attacks.They become the major targets of cyberattacks due to the increase in number of interconnections with other networks.Several research works have focused on the design of intrusion detection systems(IDS)using machine learning(ML)and deep learning(DL)models.At the same time,Blockchain(BC)technology can be applied to improve the security level.In order to resolve the security issues that exist in the critical infrastructures and ICS,this study designs a novel BC with deep learning empowered cyber-attack detection(BDLE-CAD)in critical infrastructures and ICS.The proposed BDLE-CAD technique aims to identify the existence of intrusions in the network.In addition,the presented enhanced chimp optimization based feature selection(ECOA-FS)technique is applied for the selection of optimal subset of features.Moreover,the optimal deep neural network(DNN)with search and rescue(SAR)optimizer is applied for the detection and classification of intrusions.Furthermore,a BC enabled integrity checking scheme(BEICS)has been presented to defend against the misrouting attacks.The experimental result analysis of the BDLE-CAD technique takes place and the results are inspected under varying aspects.The simulation analysis pointed out the supremacy of the BDLE-CAD technique over the recent state of art techniques with the accuy of 92.63%.

Assuring enhanced privacy violation detection model for social networks

Purpose-Virtually unlimited amounts of data collection by cybersecurity systems put people at risk of having their privacy violated.Social networks like Facebook on the Internet provide an overplus of knowledge concerning their users.Although users relish exchanging data online,only some data are meant to be interpreted by those who see value in it.It is now essential for online social network(OSN)to regulate the privacy of their users on the Internet.This paper aims to propose an efficient privacy violation detection model(EPVDM)for OSN.Design/methodology/approach-In recent months,the prominent position of both industry and academia has been dominated by privateness,its breaches and strategies to dodge privacy violations.Corporations around the world have become aware of the effects of violating privacy and its effect on them and other stakeholders.Once privacy violations are detected,they must be reported to those affected and it’s supposed to be mandatory to make them to take the next action.Although there are different approaches to detecting breaches of privacy,most strategies do not have a functioning tool that can show the values of its subject heading.An EPVDM for Facebook,based on a deep neural network,is proposed in this research paper.Findings-The main aim of EPVDM is to identify and avoid potential privacy breaches on Facebook in the future.Experimental analyses in comparison with major intrusion detection system(IDS)to detect privacy violation show that the proposed methodology is robust,precise and scalable.The chances of breaches or possibilities of privacy violations can be identified very accurately.Originality/value-All the resultant is compared with well popular methodologies like adaboost(AB),decision tree(DT),linear regression(LR),random forest(RF)and support vector machine(SVM).It’s been identified from the analysis that the proposed model outperformed the existing techniques in terms of accuracy(94%),precision(99.1%),recall(92.43%),f-score(95.43%)and violation detection rate(>98.5%).