Volume 12 | Issue 8 |

 Abstract

Cardiovascular diseases (CVDs) remain a leading cause of mortality worldwide, necessitating accurate and timely diagnosis for effective management and prevention. The integration of artificial intelligence (AI) techniques holds promise in enhancing the precision and efficiency of CVD diagnosis. This abstract outlines the framework and significance of employing AI in CVD diagnosis for a comprehensive project. The proposed project aims to develop and implement an AI-driven system for the diagnosis of cardiovascular diseases. Leveraging machine learning algorithms, particularly deep learning models, the system will analyze diverse patient data, including medical history, vital signs, imaging results, and genetic markers. Through the integration of these heterogeneous data sources, the AI model will learn complex patterns and relationships indicative of CVD presence, progression, and risk factors. Furthermore, the project will emphasize interpretability and transparency, providing clinicians with insights into the decision-making process of the AI model. The deployment of the AI-driven CVD diagnosis system in clinical settings has the potential to revolutionize cardiovascular healthcare delivery.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Integration of Artificial Intelligence (AI), AI-driven CVD diagnosis system

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The demand for authentic medicinal plants is on the rise, necessitating robust methods to ensure their integrity throughout the supply chain. In this paper, we present a novel approach leveraging machinelearning, specifically YOLOv8, for the precise identification of medicinal plants. Our methodology involves the curation of a diverse dataset comprising 30 distinct species of medicinal plants, which is used for rigorous training and testing of the model. The developed web application seamlessly integrates HTML, CSS, Bootstrap, JavaScript, React.JS, and Flask, offering a user-friendly interface for plant identification. Through comprehensive evaluation, our model demonstrates commendable performance metrics, contributing significantly to the authentication and preservation of medicinal plant integrity in the supply chain. This research not only addresses existing challenges but also paves the way for future advancements in leveraging machine learning for plant identification and supply chain management

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

MEDICINAL PLANT IDENTIFICATION USING DEEP LEARNING

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The Animal Protection System with Vibration Fencing and Solar Power Integration addresses the challenge of safeguarding animals from potential harm caused by traditional electric fencing on farms. This innovative system employs ultrasonic sensors to detect animal proximity and utilizes a vibrator motor to produce fence vibrations, effectively deterring animals without causing injury. The system is equipped with dual power sources of energy and a conventional power supply to ensure continuous functionality. In solar energy, the vibrator motor is powered sustainably, promoting environmental efficiency. The sensors trigger the system, activating the appropriate power source and the vibrator motor. As a result, animals receive a non-harmful stimulus, prompting them to retreat from the fence. This eco-friendly and human approach offers a viable alternative to conventional electric fencing, contributing to animal welfare and crop protection in agricultural settings

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

non-harmful, crop protection, Dual power sources, Agricultural Innovation, Ultrasonic sensor.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

In an era of unprecedented global connectivity, effective communication across diverse linguistic landscapes is pivotal. This project, representing a sophisticated translation application, emerges as a beacon, fostering cross-cultural understanding by seamlessly bridging language barrier. A novel translation application that integrates cutting-edge technologies to provide comprehensive translation services across text, audio, and sign language. Utilizing computer vision for accurate sign language detection, the application employs the Hugging Face mBART model for sophisticated language processing. It leverages Google Cloud Services to ensure robust and scalable translations, supporting numerous languages with high accuracy. The user interface, built with Python Streamlit, offers an intuitive and interactive experience, making it accessible to users with varied technical backgrounds. This integrated solution aims to facilitate seamless communication across different modalities, enhancing accessibility and fostering global connections.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Sign language translation, multi-language translation, computer vision, Hugging Face mBART model

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The Woven Design File Transfer Utility integrates LPC2148 and Node MCU for seamless file transfer between embedded systems and IoT networks. It ensures efficient, secure data transmission with real-time monitoring and control. Robust error handling safeguards data integrity, while scalability and flexibility cater to diverse applications. Compatibility and interoperability enable seamless integration, optimizing performance. User-friendly interfaces empower effortless configuration and management. Meticulous documentation demonstrates reliability across domains, unlocking potential in industrial automation, and beyond

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

LPC2148 , NodeMCU,-DWIN HMI

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

An This drone-based surveillance system proposes a groundbreaking solution to challenges in flood management by introducing proactive monitoring and real-time data dissemination. Equipped with high-resolution cameras and GPS capabilities, specialized drones continuously monitor flood affected areas, providing rescue teams with vital information on precise flood levels and the exact GPS locations of individuals in distress. This innovation empowers rescue teams to make informed, data-driven decisions, optimizing responses based on the severity of the situation. This solution distinguishes itself through its dynamic adaptability. In high-flood scenarios, the system recommends deploying boats for evacuation, while in low-flood situations, alternative rescue methods are employed. The ability to dynamically adjust the number of rescue team members based on real-time population data minimizes response time, reducing the risk of casualties among both flood victims and rescue teams. This comprehensive and proactive approach transforms the traditional reactive model, enhancing overall disaster management effectiveness and striving to diminish fatalities during flood emergencies.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Autonomous Aerial Surveillance , Flood Rescue ,Technology Integration. Disaster Management, Real-time Monitoring , Climate Resilience.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

There is a sustainable worldwide effect, both in terms of disease and death, that is caused by pneumonia, which is a disorder that is easily affects the children and old age people. In recent days, pneumonia has easily affect the people because of pollution, increase in population and unhygienic conditions of living. It is a respiratory infection caused by bacteria and virus that affects the lungs. Many developed and developing nations also affected by pneumonia. Pneumonia can be identified by three different ways like CT scans, US scans and X-ray images. Due to the advancement in image recognition and deep learning technologies, computed vision devices predict infected and uninfected lung images with more accuracy .To make it success in rural areas, we employ a light weighted model to run through low-cost resource constraint devices and achieved remarkable training accuracy of 98.75%. Accuracy analyzed with three different learning rates and also compared with VGG-16, Mask-RCNN, and DenseNet121.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Pneumonia, Respiratory infection, Light weight CNN, X-Ray, VGG-16, Mask-RCNN, DenseNet121.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

In urban settings, efficient public transportation systems are essential for ensuring smooth commuting experiences for residents. Central to this is the management of bus passes, which traditionally involves cumbersome processes such as manual application submissions and in-person renewals, often leading to long queues and administrative bottlenecks. To address these challenges, this project proposes the development of an Virtual Bus Pass Management System (VBPMS) designed to streamline the application and renewal processes, thereby enhancing user convenience and administrative efficiency. The VBPMS will offer a user-friendly online platform accessible via web browsers and mobile devices, allowing commuters to apply for new bus passes or renew existing ones from the comfort of their homes or offices. Through intuitive interfaces and step-by-step guidance, users will be able to input their personal details, upload required documents, and select pass options tailored to their specific needs. Overall, the Online Bus Pass Management System aims to revolutionize the way bus passes are administered, shifting from traditional paper-based methods to a seamless online platform that enhances accessibility, transparency, and efficiency for both commuters and administrative personnel. By eliminating the need for queuing and paperwork, the system promises to significantly enhance the commuter experience while reducing administrative burdens, ultimately contributing to the advancement of sustainable and user-centric public transportation systems.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

VIRTUAL BUS PASS MANAGEMENT SYSTEM

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The aim of this project helps to develop an application to find the type of attack that occurred on the system and to detect the intruders by using Intrusion Detection System. Previously various machine learning (ML) techniques are applied on the IDS and tried to improve the results on the detection of intruders and to increase the accuracy of the IDS. This paper has proposed an approach to develop efficient IDS by using the principal component analysis (PCA) and the random forest classification algorithm. Where the PCA will help to organise the dataset by reducing the dimensionality of the dataset and the random forest will help in classification. Results obtained states that the proposed approach works more efficiently in terms of accuracy as compared to other techniques like SVM, Naive Bayes, and Decision Tree. The IDS acts as a network level defence to secure a system. IDS mainly used for security purpose to find the threats or malicious activities and also for identifying the type of attack on the system.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

IDS-intrusion detection system, Dimensionality, Datasets, Intruders, PCA-principal component analysis, Accuracy, RFA-random forest approach, Attack, Detection, classification.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Our project revolves around the development of an advanced flood warning system aimed at significantly enhancing disaster response efforts. At its core, the system harnesses the power of machine learning to predict and mitigate the impact of potential flooding events. A pivotal aspect of our approach is the creation of a user-friendly interface accessible to the general public. Through this interface, individuals can access crucial information regarding the likelihood of flooding in their respective regions. By utilizing historical data on river flow and visualizing rainfall patterns at the sub-division level, users gain valuable insights into the potential risks they face. The methodology employed in our project places a strong emphasis on the utilization of machine learning algorithms. These algorithms analyze vast datasets to forecast future outcomes related to flooding with remarkable accuracy. Additionally, we prioritize the speed and timeliness of our predictive models, ensuring that users receive timely alerts and warnings well in advance of potential flood events. By proactively alerting communities to the possibility of flooding, our system aims to minimize the loss of life and property that often accompanies such disasters. Drawing on lessons learned from past incidents, such as the devastating floods that struck Tamil Nadu , we are committed to leveraging cutting-edge technology to create a safer and more resilient future for at-risk communities.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Flood prediction, machine learning , forecasting floods

  License

Creative Commons Attribution 4.0 and The Open Definition