Volume 12 | Issue 5 |

 Abstract

Abstract Natural remedies are more acceptable in the belief that they are safer with fewer side effects than the synthetic ones. Herbal formulations have growing demand in the world market. The objective of this work is to formulate and evaluate a cosmetic preparation polyherbal face pack made from herbal ingredients. Kaoline, tragacanth, orange peel powder, neem powder, chandan powder, aloe juice powder, turmeric powder , Fullers earth and Cicer arientinum Powder were procured from the local market in dried, powdered and then passed through sieve no 80, mixed thoroughly prepared and evaluated for its organoleptic, physico-chemical and microscopical characters . The dried powder of combined form had passable flow property which is suitable for a face pack. Herbal face packs or masks are used to stimulate blood circulation, rejuvenates and help to maintain the elasticity of the skin and remove dirt from skin pores. It is a very good attempt to establish the herbal face pack containing different powders of plants. The advantage of herbal cosmetics is their non-toxic nature, reduce the allergic reactions and time tested usefulness of many ingredients. Thus in the present work, we found good properties of the face packs and further optimization studies are required on this study to find the useful benefits of face packs on human, use as cosmetic product.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Keywords:- Cosmetic, Face Pack, Herbal, Ingredients, Natural Formulation.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

This project pioneers a novel method for human behavior recognition, introducing two innovative channel attention modules: the space-time interaction and depth separable convolution modules. Utilizing convolutional neural networks (CNNs), renowned for image and video processing, a multi-scale CNN approach segments behavior videos, applies low-rank learning for behavior information extraction, and integrates findings along the time axis for holistic comprehension. This method not only simplifies information extraction but also adapts flexibly to diverse network structures, enhancing recognition accuracy while minimizing computational complexity. Further, by amalgamating CNN, GRU, and Bidirectional algorithms, the model achieves superior accuracy with just 1000 parameters, outperforming existing algorithms. This hybrid approach optimizes training features, securing even higher accuracy in behavior recognition.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Human Behaviour, CNN

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

This project introduces a groundbreaking solution to the security vulnerabilities encountered by resource-limited Internet of Things (IoT) devices during software updates. Traditional methods are fraught with security risks and inefficiencies due to multiple data transfers. The proposed blockchain-based framework revolutionizes this process by leveraging Ciphertext-Policy Attribute-Based Encryption (CP-ABE) for cryptographic tasks, custom authorization policies for enhanced security, and smart contracts to ensure secure delivery and payment. By storing encrypted software update blocks across multiple IPFS nodes and recording their addresses on the blockchain, the system mitigates the risk of a single point of failure. This innovative approach not only guarantees secure, efficient, and auditable software updates but also significantly bolsters IoT device security compared to conventional methods.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Blockchain, Ciphertext

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Pneumonia is the leading cause of death and morbidity in children, and early and accurate diagnosis is essential for timely intervention. In our project, we solve the challenge of separating chest and lung x-rays. To achieve this, we use the power of neural network (CNN) and state-of-the-art transformers. Our method involves the use of a pre-trained CNN model and Vision Transformer to extract complex features from X-ray images, allowing us to identify transformation patterns and defect. We carefully preprocessed the children's chest X-ray image database to ensure that the information was complete and balanced. We have implemented the most efficient and effective testing methods to reduce translation time and cost and improve early detection of childhood pneumonia. Our system is accurate and effective in classifying lung diseases in children; It demonstrates the potential of deep learning and visual interpretation to improve doctors' ability to quickly and accurately diagnose life-threatening diseases.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Pneumonia , CNN, Vision Transformer, X-ray

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Since this study delves into the profound impacts of self-harm on individuals and economies, emphasizing the inadequacy of traditional statistics in tracking national trends. Introducing the innovative FAST framework, it harnesses social media data to forecast self-harm incidents. By training language models to discern mental health signals from online messages, this method transforms them into insightful time series data. Using machine learning regressors, the framework demonstrated superior forecasting accuracy. In a Thai case study, it surpassed conventional methods by over 40%. Additionally, incorporating the Decision Tree algorithm enhanced accuracy, reducing Mean Absolute Error compared to other algorithms. This research pioneers a transformative approach to predict nationwide self-harm trends and potentially forecast socioeconomic factors using social media analytics.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Self-Harm, Social Networks

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

as the project pioneers the use of open-source cyber threat intelligence (OSCTI) to bolster network security via a cybersecurity knowledge graph. This innovative graph streamlines access to a range of threat information, empowering informed decision-making. Leveraging attribution technology, the initiative detects and pinpoints advanced persistent threats (APTs) across diverse attack scenarios. Integrating cutting-edge knowledge graph technology with research on cyber threat attribution, the team introduces CSKG4APT, a robust cybersecurity platform. This platform harnesses ontology theory to craft an APT-centric knowledge graph model and deploys deep learning algorithms for knowledge extraction and updating. By introducing effective APT attack attribution techniques, the project amplifies network defense strategies, enabling proactive defense against rapidly evolving threats.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Cybersecurity, deep learning algorithms

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Generative AI transforms healthcare by enhancing medical imaging, accelerating drug discovery, and enabling personalized medicine. Challenges include ethics, biases, and interpretability. Future directions involve customized models and regulatory frameworks. Responsible adoption is crucial for realizing generative AI's potential. In the realm of medical imaging, generative AI reconstructs high-resolution images from low-quality scans, aiding radiologists in precise diagnoses. Additionally, it synthesizes realistic images of organs and tissues, providing valuable visual information. Personalized medicine optimization involves analyzing patient data to tailor interventions. Disease progression prediction and individualized drug dosages improve patient outcomes. While generative AI holds immense promise, addressing ethical concerns, mitigating biases, and ensuring interpretability is essential. Collaborative efforts and thoughtful regulation will drive responsible adoption and transformative impact in healthcare.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

generative ai, artificial intelligence, machine learning,healthcare,medical,chat gpt

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The Internet of Things (IoT) has revolutionized healthcare by facilitating the development of innovative solutions for remote health monitoring. This paper introduces an IoT-based health monitoring system tailored to address the challenges of monitoring vital health parameters, particularly in rural or remote areas. The system incorporates a comprehensive array of sensors, including the MAX30100 for Blood Pressure Monitoring (BPM) and Blood Oxygen Saturation (SpO2), the DS18B20 temperature sensor for precise temperature monitoring, a DHT11 sensor for humidity measurement, and a GPS module for accurate location tracking. These sensors are seamlessly integrated into a dedicated PCB board, optimizing space and efficiency. Data collected from the sensors are wirelessly transmitted to a centralized server through the Blynk IoT platform, enabling real- time analysis and visualization. An intuitive user interface empowers healthcare providers and patients to monitor health parameters and receive alerts promptly in case of deviations from normal values. Rigorous testing and validation ensure the system's reliability and accuracy across various environmental conditions. This IoT-based health monitoring system holds significant promise for enhancing. The Internet of Things (IoT) has revolutionized healthcare by facilitating the development of innovative solutions for remote health monitoring. This paper introduces an IoT-based health monitoring system tailored to address the challenges of monitoring vital health parameters, particularly in rural or remote areas. The system incorporates a comprehensive array of sensors, including the MAX30100 for Blood Pressure Monitoring (BPM) and Blood Oxygen Saturation (SpO2), the DS18B20 temperature sensor for precise temperature monitoring, a DHT11 sensor for humidity measurement, and a GPS module for accurate location tracking. These sensors are seamlessly integrated into a dedicated PCB board, optimizing space and efficiency. Data collected from the sensors are wirelessly transmitted to a centralized server through the Blynk IoT platform, enabling real- time analysis and visualization. An intuitive user interface empowers healthcare providers and patients to monitor health parameters and receive alerts promptly in case of deviations from normal values. Rigorous testing and validation ensure the system's reliability and accuracy across various environmental conditions. This IoT-based health monitoring system holds significant promise for enhancing.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

iot devices, internet connectivity, cloud platform, data security, mobile or web application, data analytics, alerting mechanisam, integration with electronic health records.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Sepsis is a potentially life-threatening condition that occurs when the body's response to an infection causes inflammation throughout the body. This inflammation can trigger a cascade of changes that can damage multiple organ systems, leading to organ failure and death if not treated promptly. Early recognition and aggressive treatment with antibiotics and supportive care are crucial for improving outcomes in septic patients. The " Accurate Prediction of Sepsis in ICU Patients" is a project that combines awareness and predictive modeling to address sepsis, a life-threatening condition commonly encountered in intensive care units (ICUs). This project is a robust awareness campaign designed to educate both the general public and healthcare professionals about sepsis. With focusing on generating awareness about sepsis, can lead to early detection and seeking medical help. By bringing the limelight on this disease, it can potentially save lives. Concurrently, advanced machine learning techniques, specifically random forest algorithms, are employed to construct a predictive model for sepsis. This model undergoes meticulous fine-tuning to ensure accurate identification of sepsis risk in ICU patients. It uses a dataset for training the predictive model. The integration of the Sequential Organ Failure Assessment (SOFA) score, including the quick SOFA (qSOFA) criteria, enhances predictive accuracy. The qSOFA criteria play a crucial role in rapid risk assessment for early intervention. Moreover, the project maintains a dedicated website that serves as an essential platform for sepsis education and the dissemination of the predictive model to the medical community.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Sepsis, ICU, SOFA, qSOFA, Random Forest, Predictive Modelling, Awareness Campaign

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Pure essential oils are concentrated oils generated from a variety of natural plants, flowers, plant roots, seeds, resins, plant exterior tissue, trees or shrubs, and fruit rinds. These oils are well-known among humans for their benefits to the body, skin, and spirit. These oils are also commercially employed due to their superior medicinal or odoriferous characteristics. To research extraction strategies available to extract oils from plants and trees, to come across pros and disadvantages of a few extraction methods, selection and efficiency of a single method. The method used to extract essential oil from plants is critical, as some processes employ solvents that can harm the healing benefits of plants and trees. There are different extraction procedures, but the oil's quality and production never remain consistent. The Soxhlet apparatus technique was used in this investigation because of its mild extraction conditions and low operating cost. Steam is a critical component in the oil extraction process. Extraction of essential oils using diverse methods and innovative techniques reduces the risk of losing the vital component of plants and trees, reduces chemical risk, shortens extraction time, is environmentally friendly, and improves the quality and production of essential oils.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Eucalyptus oil, Steam Distillation using Soxhlet apparatus

  License

Creative Commons Attribution 4.0 and The Open Definition