Review of an EMG-Controlled Prosthetic Arm

Amputation, especially of the upper limbs, is a condition that exists in almost all parts of the world. There are more than 110 thousand amputees in India itself. It is extremely difficult for amputees to carry out their daily activities and to deal with daily life as normal people do. The purpose of the myoelectric prosthesis is to restore the basic functions of the lost organs in the joint using neural signals produced by the muscles. Unfortunately, the use of such myosignals is complicated. In addition, once detected, it usually requires a computational force strong enough to convert it into a user-controlled signal. Its modification to the actual function of the implant is limited by a number of factors, especially those associated with the fact that each amputee has a different muscle movement. Modified artificial intelligence systems designed for pattern recognition have the potential to improve the size of implants but still fail to provide a system in which artificial arms can be controlled by brain signals.

Thermal Analysis of Realistic Breast Model With Tumor and Validation by Infrared Images

Breast thermography is an emerging adjunct tool to mammography in early breast cancer detection due to its non-invasiveness and safety. Steady-state infrared imaging proves promising in this field as it is not affected by tissue density. The main aim of the present study is to develop a computational thermal model of breast cancer using real breast surface geometry and internal tumor specification. The model depicting the thermal profile of the subject's aggressive ductal carcinoma is calibrated by variation of blood perfusion and metabolic heat generation rate. The subject's IR image is used for validation of the simulated temperature profile. The thermal breast model presented here may prove useful in monitoring the response of tumor post-chemotherapy for female subjects with similar breast cancer characteristics.

Metaheuristic Techniques of Parameter Estimation of Solar PV Cell

The performance and efficiency of a solar PV cell are greatly dependent on the precise estimation of its current-voltage (I-V) characteristic. Usually, it is very difficult to estimate accurate I-V characteristics of solar PV due to the nonlinear relation between current and voltage. Metaheuristic optimization techniques, on the other hand, are very powerful tools to obtain solutions to complex non-linear problems. Hence, this chapter presents two metaheuristic algorithms, namely particle swarm optimization (PSO) and harmony search (HS), to estimate the single-diode model parameters. The feasibility of the metaheuristic algorithms is demonstrated for a solar cell and its extension to a photovoltaic solar module, and the results are compared with the numerical method, namely the Newton Raphson method (NRM), in terms of the solution accuracy, consistency, absolute maximum power error, and computation efficiency. The results show that the metaheuristic algorithms were indeed capable of obtaining higher quality solutions efficiently in the parameter estimation problem.

A Compact Antenna Design With High Gain for Wireless Energy Harvesting

A compact rectangular slotted antenna fed through coplanar waveguide for rectenna system is proposed in the application of radio frequency (RF) energy harvesting at center frequency of 2.45 GHz in the wireless local area network (WLAN) band. Three unequal widths of rectangular slots with equal distance have been created step by step to maximize the peak gain to 3.6 dB of the antenna. Radiation plot of the proposed antenna has been depicted to be omnidirectional for RF energy harvesting with maximum radiation efficiency characteristics. The dimension of the antenna is reduced up to 28 × 17 mm2 with better reflection coefficient of -34.6dB.

Reconstruction of 2D Images Into 3D

3D reconstruction is a long-standing complication when comes to testing happening from decades from machine learning, computer graphics, and computer perspective environments. Using CNN for the reconstruction of the 3D image has enchanted growing attentiveness and shown spectacular execution. Emerging in the new era of abrupt development, this chapter lays out an in-depth study of the latest developments in the field. Its focuses on activities that use in-depth learning strategies for measuring the 3D status of common things from one or more RGB images. It organizes based on literature in the layout presentations, network structures, and training methods they use. As the survey was conducted for methods of reconstructing common objects, this chapter also evaluates some of the latest efforts that emphasize particular categories of an object such as the shape of the human body and face. This provides an examination and correspondence of the execution of some important papers, summarizing some open-ended issues in the field, and exploring encouraging indications for subsequent research.

Machine Learning Approaches for Cardiovascular Disease Prediction

In a human body, the heart is the second primary organ after the brain. It causes either a long-term impairment or death of a person if suffering from a cardiovascular disease. In medical science, a proper medical analysis and examination of a cardiovascular disease is very crucial, convincing, and sophisticated task for saving a human life. Data analytics rises because of the absence of sufficient practical tools for exploring the trends and unknown relationships in e-health records. It predicts and achieves information which can ease the diagnosis. This survey examines cardiovascular disease prediction systems developed by different researchers. It also reviews the trend of machine learning approaches used in the past decade with results. Related studies comprise the performance of various classifiers on distinct datasets.

Comparison of Machine Learning Algorithms for Cardiovascular Disease Prediction

Cardiovascular disease prediction is a research field of healthcare which depends on a large volume of data for making effective and accurate predictions. These predictions can be more effective and accurate when used with machine learning algorithms because it can disclose all the concealed facts which are helpful in making decisions. The processing capabilities of machine learning algorithms are also very fast which is almost infeasible for human beings. Therefore, the work presented in this research focuses on identifying the best machine learning algorithm by comparing their performances for predicting cardiovascular diseases in a reasonable time. The machine learning algorithms which have been used in the presented work are naïve Bayes, support vector machine, k-nearest neighbors, and random forest. The dataset which has been utilized for this comparison is taken from the University of California, Irvine (UCI) machine learning repository named “Heart Disease Data Set.”

Reliability of Smart Grid Including Cyber Impact

Smart grid has changed power systems and their reliability concerns. Along with that, cyber security issues are also introduced due to the use of intelligent electronic devices (IEDs), wireless sensory network (WSN), and internet of things (IoT) for two-way communication. This chapter presents a review of different methods used from 2010 to 2020 focusing on citation as the main criteria for reliability assessment of smart grids and proposals to improve reliability when it comes to assessing a practical transmission system. It shows that evolutionary techniques are the latest trend for smart grid security.

Catch Me If You Can Game as Well as Packaging System Efficient Design Using Arduino UNO

This chapter is about a basic power-efficient object dropping game named “Catch Me If You Can,” which works on the Arduino platform. A dropping mechanism is developed using the servo motors interfaced to Arduino. The mechanism includes an object holder attached to the servomotor and a loading tube. The dropping of the objects is controlled by any wireless Bluetooth controlling device like a mobile phone or joystick and by the keypad interface installed in the design. The game is about catching the objects dropping randomly as a challenge which is controlled by the operator. The overall simulated design can be done in EasyEDA platform. The overall game can be controlled by an app which is designed by MIT App Inventor. This game can be implemented in Amusement parks, exhibitions, kid schools, and shopping malls. Besides this entertainment aspect, commercially it works as a small-scale product packaging system by involving DC motors, which are needed in moving packaging belts. This mechanism is efficient in packaging products in some specific count.

Machine Learning Based Intrusion Detection System for Denial of Service Attack

Machine learning-based intrusion detection system (IDS) is a research field of network security which depends on the effective and accurate training of models. The models of IDS must be trained with new attacks periodically; therefore, it can detect any security violations in the network. One of most frequent security violations that occurs in the network is denial of service (DoS) attack. Therefore, training of IDS models with latest DoS attack instances is required. The training of IDS models can be more effective when it is performed with the help of machine learning algorithms because the processing capabilities of machine learning algorithms are very fast. Therefore, the work presented in this chapter focuses on building a model of machine learning-based intrusion detection system for denial of service attack. Building a model of IDS requires sample dataset and tools. The sample dataset which is used in this research is NSL-KDD, while WEKA is used as a tool to perform all the experiments.