Browse Articles
Discover research articles across all indexed journals
Identification and Categorization of Skin Cancer Using Deep Learning
The second leading cause of death worldwide is cancer. By and large, the commonness of disease has really expanded; simply in the US alone, roughly 1,665,540 individuals experienced malignant growth, and 585,720 of them passed on because of this sickness by 2014. Consequently, disease is a difficult issue influencing the strength of every single human culture. Sadly, it is an assortment sickness at the tissue level and this assortment is difficult for its particular conclusion, trailed by viability of therapy. In men, the most noteworthy rates of malignant growth types happen in the prostate, lung and bronchus, colon and rectum, and urinary bladder, separately. In ladies, malignant growth commonness is most elevated in the bosom, lung and bronchus, colon and rectum, uterine corpus and thyroid, separately. This information demonstrates that prostate and bosom disease comprises a significant piece of malignant growth in people, separately. For youngsters, the most elevated rate kinds of malignant growth illness are blood disease, and tumours connected with the mind and lymph hubs, separately. Disease happens by a progression of progressive transformations in qualities so these transformations change cell capabilities. Cancer cells and gene mutations are clearly caused by chemical compounds. What's more, smoking includes a few cancer-causing substances intensifies that lead to cellular breakdown in the lungs. Strangely, natural compound substances with cancer-causing properties impact straightforwardly or in a roundabout way the cytoplasm and core of cells, and lead to hereditary problems and quality transformations. Infections, microbes and radiation beams are other carcinogenesis factors, involving around 7% of all malignant growths. Cancer generally results in the dysfunction of essential genes and disrupts the relationships between cells. This aggravation is emotional in the cell cycle, and prompts unusual expansion. Proto-oncogenes are answerable for cell division and development under typical condition; however become oncogenes during hereditary transformation, which are generally perilous for cell presence. Also, the absence of growth silencer qualities triggers uncontrolled cells division. Ordinarily, fix qualities mean protein and catalysts that have fixing properties and in excess of 30 kinds of recognized fix proteins. Eliminating uracil from DNA sidesteps the DNA harm and eliminates the principal DNA sores prompted by bright light, which are basically the elements of fix qualities to fix DNA 18 effectively. In the study of cell fate and epigenetic modifications like DNA methylation, histone modifications, and nucleosome position, which play important roles in the formation of cancer, epigenetics is a dynamic field. Disease cells are portrayed by an immense decrease in DNA methylation (around 5-6% decrease in the aggregate sum of 5-methyl cytosine). In cancer cells, the majority of histone modifications consist of a reduction in mono-acetylated H4K16. Cancer is linked to all families of chromatin-modifying proteins, but the molecular mechanisms underlying their functions are typically unknown. In this review, we evaluated malignant growth according to the point of view of the sub-atomic level, to draw a nearer check this illness out.
Voice Recognition Based Vital Parameter Monitoring Patient BED
In healthcare protocols continuous monitoring of vital parameters plays a crucial role in patient care. Integrating advanced technologies for both monitoring and patient comfort enhances the efficiency of healthcare delivery. This paper presents the design and development of a novel patient bed system equipped with vital parameter monitoring capabilities and adjustable head position controlled through voice commands. The proposed system leverages state-of-the-art sensors to monitor vital parameters such as heart rate, blood pressure, oxygen saturation, and temperature. These sensors are seamlessly integrated into the bed, providing non-intrusive and continuous monitoring of the patient's health status. Real-time data acquisition and processing algorithms ensure accurate and timely assessment of the patient's condition. Furthermore, the patient bed is designed with an adjustable head position feature, allowing for optimal comfort and support. Unlike conventional manual or automated adjustment mechanisms, our system employs voice control technology for intuitive and hands-free operation. Patients can simply issue verbal commands to adjust the head position according to their preferences, enhancing overall comfort and facilitating better rest.
Survey on Power and Area efficient FIR Filter Architecture in Digital Encephalography System
In the process of Electroencephalogram (EEG) acquisition, the electrical signals from the brain are contaminated by numerous noise sources and artifacts. American Clinical Neurophysiology Society recommends digital filtering of acquired EEG signals with an upper cut-off frequency of 70 Hz before display in digital encephalography systems. We propose a linear symmetrical FIR filter architecture for filtering EEG signals using approximate computation techniques. Approximate computation techniques result in lower hardware resources and lower power consumption with some compromise in quality. Software tools used for this study include MATLAB (and its FDA tool) and Xilinx ISE 14.7. The chosen target platform is the Spartan-3e starter FPGA board. We propose the architecture for an Approximate Dadda Tree Multiplier. The proposed approximate multiplier consumes 25% lesser slices and 35.18% lower dynamic power than state-of-the-art approximate multipliers. The proposed FIR filter consumes 19.77% lesser LUTs and 34.97% lower power than state-of-the-art symmetric FIR filter architectures.
Diagnostic Wearable Sensing Device for Blood Leak Detection during Hemodialysis Using an Array Sensing Patch
One important medical procedure that necessitates puncturing the skin is hemodialysis (HD). But there is a serious chance that blood will spill from a dislodged needle, endangering the lives of the patients. Numerous devices that use optical or electrical methods to identify blood loss during HD have been proposed in the past. Sadly, multi-bed monitoring was not feasible with these techniques, which were limited to single-point detection only. An array-sensing patch has been proposed as a novel wearable device to monitor blood loss during HD in response to these restrictions. Risk levels can be measured and transmitted using the array sensing patch in conjunction with a wireless module and mapping circuit. This has the potential to greatly improve health care worker productivity and eliminate erroneous risk levels. In addition to lowering false alarms, this can greatly improve health care personnel' productivity. Extensive experimentation results show that up to 0.1 mL of blood loss can be detected a teach spot in the sensing array. In addition, the array sensing patch includes an 8-hour risk level monitoring system, which notifies medical staff of relevant patient danger. A more effective and efficient means of monitoring blood leakage and potentially saving lives, the suggested device is a game-changer in the field of hemodialysis.
Environmental Impact Assessment of Chiplet-Based VLSI
In this paper, we explore the revolutionary shift in VLSI design brought about by chiplet-based architectures, which are superior to the older, single-piece designs in many ways. This type of architecture not only boosts performance and allows for greater system expansion but also plays a significant role in promoting environmental sustainability. Chiplet-based designs lead to more resource-efficient, less wasteful, and more energy-efficient semiconductor technologies, steering the industry towards a more environmentally friendly path. The adoption of eco-chips enhances these environmental benefits, leading to a notable decrease in the carbon emissions associated with electronic devices and systems. This shift is pivotal in achieving the twin goals of advancing technology and upholding environmental stewardship in VLSI. We introduce HI as a strategy for achieving sustainable computing, creating chiplet-based systems that have a smaller carbon footprint than traditional System on Chips (SoCs). We also present ECO-CHIP, a tool we developed for estimating the carbon footprint of these systems, accounting for both their construction and operational phases, including the carbon costs of advanced packaging. In Section VI, we illustrate how ECO-CHIP can be used for breaking down systems into smaller parts and for exploring different design options, and we have integrated it with other tools for estimating the costs of chiplet-based designs. ECO-CHIP is freely available as open-source software and can be found in an anonymous repository [21]. We contend that ECO-CHIP will facilitate the creation of more environmentally friendly design practices for the next generation of complex systems.
IoT Enabled 3D Printer Device for Bio models
IoT and embedded electronics is getting popular due to its versatile applications in various fields for domestic, commercial and industrial usage. However electronics or electrical systems that are primarily built using electronics, needs enclosures to protect it from the environment as well as for maintaining its privacy concerns. 3D printers are valuable to rapidly prototype/fabricate/build enclosures for custom applications. The steps, involved in rapid prototyping using 3D printers start from understanding the specification and measurement requirements e.g. colour, look, dimensions, and tolerances. Cura is a popular slicing software used in 3D printing that converts your 3D models (STL files) into a format (G-code) that your 3D printer can understand. Integrating IoT (Internet of Things) with 3D printing using Modified OctoPrint. It can greatly enhance the functionality and accessibility of your 3D printing setup. OctoPrint is an open-source software that allows you to control and monitor your 3D printer remotely via a web interface. By adding IoT capabilities, you can further automate and monitor your printing processes, making them more efficient and comfort.
Feed Forward Neural Network With Column-Wise Matrix–Vector Multiplication on FPGAs
This article presents a reconfigurable accelerator for recurrent neural networks with fine grained column wise matrix vector multiplication (RENOWN).We propose a novel latency hiding architecture for recurrent neural network accelerator using column wise matrix vector multiplication instead of the state of the art row wise operation. This hardware (HW) architecture can eliminate data dependencies to improve the throughput of RNN inference systems. Besides, we introduce a configurable checkerboard tiling strategy which allows large weight matrices, while incorporating various configurations of element-based parallelism (EP) and vector-based parallelism (VP). These optimizations improve the exploitation of parallelism to increase HW utilization and enhance system throughput. Evaluation results show that our design can achieve over 29.6 tera operations per second (TOPS) which would be among the highest for field-programmable gate array (FPGA)-based RNN designs. Compared to state-of-the-art accelerators on FPGAs, our design achieves 3.7–14.8 times better performance and has the highest HW.
Vehicle Headlight Automatic Bright and Dim Control System
Headlights of vehicles pose a great danger during night driving. The drivers of most vehicles use high, bright beam while driving at night. This causes a discomfort to the person travelling from the opposite direction and therefore experiences a sudden glare for a short period of time. This is caused due to the high intense headlight beam from the other vehicle coming towards the one from the opposite direction. In this paper, an automatic headlight dimmer which uses a sensor has been designed to dim the headlight of on-coming vehicles to avoid human eye effects. This automatically switched the high beam into low beam, therefore reducing the glare effect by sensing the light intensity value of approaching vehicle. Road accidents are a serious cause of concern and glare from headlamps of vehicles during night time create havoc for the drivers making driving really an uphill task and due to the undesirable situation, drivers meet with accidents. It’s a very hectic job for the drivers to manually control the beam of the headlamps during night from time to time when an oncoming vehicle is located within 150 meters of the other vehicle. So, it’s high time when a safety control unit should be installed in vehicles that can automatically dim the headlamps by detecting their intensity which, otherwise, might cause a serious trouble for the drivers. Our work proposes a cost effective and useful control unit which will automatically detect the excessive glare from headlamps and will give a signal to the oncoming vehicle causing the trouble to dim the lamps. As such, the driver’s unnecessary load will be reduced and driving will become safer and smoother during night.
Customized Tricycle for Patient with Disabilities
Mobility of physically disabled persons is a concerning social issue nowadays. The basic Tricycle is a three- wheeled design, pedaled by disabled persons in the side and seat in the middle for sitting arrangement. They use only one hand to steer the handle because other-hand is used to rotate the pedal. Our aim is to design a low-cost tricycle that will allows person to accelerate and steer the tricycle from a single place. The way the Bike Chair works is the back is essentially a regular bicycle, but the front turns it into a tricycle, and sort-of makes it a wheelchair and bicycle combo. The disabled passenger sits on the front two wheels, as the driver uses their front wheels to steer the bike. The passenger has buckles and strap to secure the min place and top even them from falling out of the chair during the bike-ride. Furthermore, the tricycle's compact size, maneuverability, and zero emissions make it an attractive option for reducing traffic congestion and pollution in densely populated urban areas. Future research directions include optimization of battery and motor technologies for enhanced performance and efficiency, as well as integration of smart features for improved user experience and safety.
Enhancing Security with Arduino uno-Powered Patrolling Robot
The development of a patrol robot using an Arduino Uno, ultrasonic sensors, PIR sensors, and the ESP32-CAM module represents an innovative approach to surveillance and automation. This autonomous robot is designed to navigate and secure designated areas efficiently. The ultrasonic sensors enable obstacle avoidance, while the PIR sensors detect motion accurately. The ESP32-CAM module provides real-time video surveillance capabilities. The robot is controlled and monitored remotely through a user-friendly interface, making it ideal for applications such as home security, monitoring, and automated patrolling. Future enhancements may involve integrating advanced navigation algorithms, and cloud-based data analysis to further expand its capabilities and adaptability.
Lora Based Coal Mine Safety Monitoring and Security System Using Smart Helmet
The process of removing coal from the earth is called coal mining. Coal is used as fuel by the steel and cement industries to extract iron from iron ore and to make cement. The category that includes the underground mining business includes all factors that require routine monitoring, including methane gas, high temperatures, fire accidents, and so on. The safe production level of coal mines is still low, and frequent tragedies there result in significant losses of life and property. It is crucial to keep an eye on the working conditions in coal mines since the variety of work done there and the complexity of the mine environment are the main causes of disasters. The coal mine based on LoRa Long-Range Wide Area Network (LoRaWAN) technology is used by the Safety Monitoring and Security System to improve security and safety in coal mines. This system incorporates cutting-edge sensors to continuously monitor the surrounding environment, gas concentrations, and worker welfare. LoRa enables seamless communication and quick identification of potential risks because to its long range capabilities.
IOT Based Water Quality Detection for Sugar Factory
This paper proposes an IoT-based water quality detection system tailored for sugar factories. Employing sensors and microcontrollers, it continuously monitors crucial parameters like pH, turbidity, and dissolved oxygen levels. Data is transmitted wirelessly to a central server for real-time analysis and alerts. This system enhances operational efficiency by enabling timely interventions to maintain water quality standards, ensuring optimal production processes.
Building a Full Stack Solution for Seamless Device Connectivity and Data Management
This paper involves developing a full-stack IoT application with a cloud server. The project encompasses defining the project scope, selecting suitable hardware, developing firmware/software for IoT devices, setting up a cloud server with IoT services, configuring data ingestion and storage, building a backend application for data processing, creating a frontend UI/dashboard, establishing bidirectional communication, conducting testing, deploying the applications, and implementing continuous monitoring and optimization. By following these steps, the project aims to create a comprehensive IoT solution that collects, analyzes, and visualizes data from connected devices, enabling remote monitoring and control.
Artificial Intelligence Based CCTV-Accident Detection System
Urban infrastructure and public safety are severely hampered by traffic congestion and accidents. This research suggests an enhanced traffic light control system that makes use of deep learning methods and IoT integration to address these problems. The system computes vehicle density in real-time and indicates smart light conditions at traffic signals by combining the object identification algorithm of YOLOv3 with Internet of Things sensors. The secret to the system is the use of YOLOv3 for precise vehicle identification and categorization in video feeds recorded by security cameras. The technology can effectively identify cars, collisions, and violent incidents on the road by utilising YOLOv3. In order to enable quick emergency response, the system sends notification alerts via GSM connection to neighbouring police stations and hospitals upon detection. In addition, the suggested system uses Internet of Things devices to calculate the number of vehicles at crossings. Traffic signal timings can be dynamically adjusted thanks to this real-time data on vehicle flow, which improves traffic flow and lessens congestion. IoT integration also makes it possible for traffic signals to have smart light indication, which adjusts signal timings in response to current traffic circumstances to improve efficiency and safety.
Generation of Power from Footpath Using IOT
The present scenario, the requirement for energy has been increasing in a distressing rate and the availability of the energy resources are not abundant for sustainable development and the need of the hour is to establish an economical pollution free inexhaustible energy resource to compensate the increasing demands. Since, walking is the most common activity in human day to day life, whenever a person walks, he exhausts energy to the ground which goes as a waste. In order to preserve and make use of this energy we are converting it into electrical energy using piezoelectric effect. Piezoelectric effect is utilized by piezoelectric sensor producing output energy in the form of AC voltage. To produce energy in a large extent, in this model implementation of the piezoelectric sensor network along the footpath is carried out which is given to the lane of streetlights along with it the smart shoe is implemented for a small production that consists GPS tracking. In this proposed work model, GPS tracking has been added and also streetlights switching technique comparing with existing model
Secured and Privacy in Exam Cell Management System Using Blockchain Mechanism with Advanced Encryption Standard (AES) Algorithm
This study proposes a novel approach to enhance the security and privacy of Exam Cell Management Systems (ECMS) by integrating blockchain technology with the Advanced Encryption Standard (AES) algorithm. Traditional systems face challenges such as data breaches and unauthorized access, necessitating innovative solutions. Our proposed system utilizes blockchain's decentralized and immutable nature to securely record and verify examination-related transactions. Each transaction is encrypted using AES before being added to the blockchain, ensuring confidentiality and tamper-proofing. The decentralized nature of blockchain enhances transparency and accountability, while AES encryption provides robust data protection. Experimental results demonstrate significant improvements in data security, privacy, and integrity compared to traditional systems. This approach offers a promising solution for educational institutions to mitigate the risks associated with data breaches and unauthorized access, thereby ensuring the integrity and trustworthiness of the examination process.
Peer-to-Peer Network Node Distribution Analysis Using Block chain
This project uses block chain technology to propose a decentralized access control method for anonymous authentication and secure cloud data storage. The suggested plan improves privacy by guaranteeing that users can authenticate themselves to the cloud without disclosing who they are. Block chain integration improves system security by removing single points of failure and enabling robust, decentralized access management. The plan handles user revocation, enables fine-grained access control, and successfully thwarts replay assaults. Users have similar communication, processing, and storage overheads to centralized methods when creating, editing, and reading material that is stored on the cloud. Our method is decentralized, which sets it apart from conventional models and makes it a useful and effective way to store data securely on the cloud for applications that need a high degree of trust.
YouTube Spam Comment Detection
This project presents a successful implementation of an advanced YouTube spam comment detection system, significantly reducing spam across the platform. Utilizing NLP, machine learning, and community reporting, the system offers a dynamic approach to content moderation. The report delves into ethical considerations, highlighting the complexity of balancing spam combat with user freedom. Serving as a notable case study, it provides a valuable blueprint for addressing spam-related challenges on expanding social media platforms, enhancing the overall user experience.
E-Commerce Mobile Applications for Visually Impaired Persons
A Mobile app design that allows visually impaired and blind users to access and shop without any assistance. The system gives voice instruction to users and an input is requested from the user. The Voice Command-Based Product Information System for Blind People is an innovative solution designed to empower visually impaired individuals with easy access to product information.
The Efficient Employee Attendance Tracking and Payroll Calculation with Geo-Fencing Technology
The Efficient Employee Attendance Tracking and Payroll Calculation with Geo-fencing Technology project aims to simplify the process of managing employee attendance and payroll using advanced technology. The proposed system utilizes the latest geo-fencing technology, Google Map API or JS geolocation API for location tracking, and PHP and MYSQL for backend development. The system offers separate dashboards for the admin and the employee, each with unique features and functionalities.