IFIP Open Access Digital Library
Not a member yet
22614 research outputs found
Sort by
Electroencephalogram Based Stress Detection Using Machine Learning
No full textPart 3: Applications of MLInternational audienceThis study analyses the topic of stress, highlighting the vital necessity of efficient coping techniques and mentality-based approaches to identify different stress states in a timely manner. The study focuses on the intricate function of electroencephalogram (EEG) signals in detecting stress for medical professionals’ diagnosis so that treatment of disorders linked to stress could be treated at the earliest. The suggested method begins with preprocessing, feature extraction and further proceeds with development of ensemble model. This work investigates the identification of stress using EEG signals. The ensemble model achieves an amazing 98% accuracy on scaled data after undergoing thorough testing. The study emphasizes the critical importance of stress identification and presents ensemble technique for sophisticated EEG-based stress identification. This underscores the importance of stress identification and how it can affect the way a person perceives and handles stress, which will ultimately lead to an improved quality of life
Healthify App Using Blockchain with Cloud
No full textPart 1: Applications of AI/ML in KDM, Cloud Computing & SecurityInternational audienceIn the modern healthcare industry, hospitals still use outdated data management systems for patient data, and the way patient health records are kept and safeguarded does not reflect our technological advancements in this field. Blockchain technology can completely transform the healthcare sector by enhancing the security and storage of medical data and records. This project uses cloud computing to implement blockchain technology in the health sector. If we have an internet connection, cloud storage enables data access on any device, at any time, and from any location. In the healthcare industry, blockchain offers advantages like accurate patient care, and real-time data access. This project proposes hosting a blockchain code on a cloud network to develop a prototype app for the healthcare sector. By utilizing cloud infrastructure, the solution can adapt to the evolving needs of the healthcare industry while meeting regulatory requirements
Integration of Raman Spectroscopy, On-Line Microscopic Imaging and Deep Learning-Based Image Analysis for Real-Time Monitoring of Cell Culture Process
No full textPart 5: Perceptual IntelligenceInternational audienceTraditionally, condition monitoring of mammalian cell culture processes is based on sampling and off-line analysis, which is labour intensive, time consuming, and causes time delays. In this work, in situ microscope and on-line Raman spectroscopy are investigated for simultaneous measurement of multiple properties of the cell growth state and biochemical indices of suspended animal cells. The focus is on investigation of deep learning-based Mask R-CNN algorithm for image analysis. The model is trained by 184 images with 183,040 cells using data augmentation methods and transfer learning technique. Mask R-CNN segments the clustered cells more effectively than the conventional one combining edge detection, intensity thresholding, and advanced watershed method. The evolution of geometrical features of cells is further analyzed, including equivalent diameter, circularity, and aspect ratio. It demonstrates the great potential of deep learning in the analysis of on-line images for control of the cell culture process
Research on Object Detection for Intelligent Sensing of Navigation Mark in Yangtze River
No full textPart 5: Perceptual IntelligenceInternational audienceThe maintenance and management of navigation marks are essential for ensuring the safety of transportation on the Yangtze River. Considering the current inspection and management approaches, this paper introduces an intelligent method for inspecting inland river navigation marks using Unmanned Aerial Vehicles (UAVs). The method enables real-time monitoring of navigation marks using UAV video inspection. A UAV data acquisition platform captures video images of these marks. We have developed the ED-YOLOv5s object detection algorithm to detect and classify navigation marks. Building on this, the system can automatically assess the light quality and status of navigation marks at night. The ED-YOLOv5s algorithm is an enhancement of the YOLOv5s model, incorporating the ECA mechanism and DFFN structure, which are based on the ResNet principle. This modification enhances the model’s capability for network feature fusion. Experimental results indicate improvements in navigation mark detection with the ED-YOLOv5s. Although precision decreased by 1.76% when compared to the YOLOv5s model, recall and [email protected] increased by 3.59% and 2.97%, respectively. The detection results for light quality state from video images of navigation marks at night accurately reflect actual conditions. We have developed an intelligent sensing scheme for navigation marks on the Yangtze River based on the improved model. This scheme has been implemented in the Yichang section of the Yangtze River, significantly reducing the cost of daily inspections, enhancing cruise monitoring effectiveness, facilitating intelligent maintenance decisions for navigation marks, and further ensuring the navigational safety of the Yangtze River
End-To-End Control of a Quadrotor Using Gaussian Ensemble Model-Based Reinforcement Learning
No full textPart 1: Machine LearningInternational audienceIn recent years, the rapid development of deep reinforcement learning has provided a new way to solve the robot control problems. However, the low sample efficiency and slow convergence speed of deep reinforcement learning have become one of the obstacles when transitioning from simulation to the real world. In this paper, we propose a quadrotor control policy using Gaussian ensemble model-based reinforcement learning. Unlike traditional control methods, this method uses an actor-critic deep neural network which is updated with a reward function to achieve end-to-end control of the quadrotor by establishing a mapping between the quadrotor's states and motor control signals. Additionally, we improve sample efficiency by constructing an ensemble model following a Gaussian normal distribution, which differs from conventional model-free RL methods. The environment model is trained using data from the agent's interaction with the real environment and reduces the number of interactions with the real environment by generating simulated data. The approach is evaluated in the AirSim which is a high-fidelity visual and physical simulator. The results show that the proposed approach improves the sample efficiency, eliminates oscillations and steady error, and demonstrates robustness to external disturbances
A Framework of Reinforcement Learning for Truncated Lévy Flight Exploratory
No full textPart 1: Machine LearningInternational audienceDeep reinforcement learning (DRL) still explores insufficiently when dealing complex tasks with high-dimensional and large state spaces. Therefore, developing better exploration strategies is still one of the important tasks in reinforcement learning. The paper introduces a new exploration strategy ATLF (Adjustment of Truncated Lévy Flight exploration framework, ATLF) which augments the existing exploration mode with Lévy flight, making action selection more stochastic to boost exploration. The ALTF framework is combined with discrete-space algorithm DQN and continuous space-algorithm SAC to handle reinforcement learning tasks. Compared with a variety of reinforcement learning algorithms on OpenAI gym environments such as MountainCar-v0 and Walker2d-v2, the result shows that our algorithm has better exploration ability than vanilla DQN or SAC, obtaining higher overall rewards, and is less likely to fall into local optimization, and is more stable. Additionally, the result shows that the ALTF is highly compatible with existing deep reinforcement learning algorithms
Evolving Cybersecurity Challenges in the Age of AI-Powered Chatbots: A Comprehensive Review
No full textPart 4: Cybersecurity and SafetyInternational audienceIn today’s world of super-digitization and dynamic transformation, the Artificial Intelligence (AI)-based chatbot is a revolutionary stride in technology. This innovative chatbot type is characterized by unmatched technological advancement across the globe in terms of efficiency and interactivity within several fields. Moving from simple automated scripts to sophisticated natural language processing systems like Chat GPT stands as a remarkable leap in conversational technology. However, this development conveys distinct computer security threats, as AI upholds a dual role in strengthening and weakening digital security. This study explores the conceivable risks that come with AI-Chatbots, as well as the appropriate mitigation strategies. This paper outlines the current risks, threats, and consequences posed by these digital assistants, and discusses useful strategies and methods to mitigate these risks and protect personal data and sensitive information. Combining current research and perspectives, to provide a balanced view of the challenges and opportunities that AI-Chatbots present in the cybersecurity domain. Establishing a guide to the development of my PhD proposal, an Intelligent Chatbot that combines the most advanced LLMs with its databases, in a safe, functional, and user-friendly user interface
synple: A Platform for Privacy Preserving Synthetic Patient Data Generation
No full textPart 3: Human-Centric Biomedical SystemsInternational audiencePatient data collection is often constrained by accessibility, privacy, and confidentiality issues in healthcare research. To overcome this problem, the generation of synthetic data has been proposed. Nevertheless, existing synthetic patient generators do not cover European populations, nor do they offer flexibility in the process of generating data and building a biographical profile. In this paper, we introduce synple, a tool for synthetic patient data generation that mirrors the demographic characteristics of Portugal and Spain while adhering to GDPR and HIPAA regulations. Our platform produces comprehensive patient profiles, including both biological and biographical details, life narratives, and facial images, facilitated through an intuitive web interface powered by advanced generation algorithms. Additionally, the platform incorporates a validation feature, enabling users to assess the quality and consistency of the generated synthetic data. We demonstrate our system’s potential for enhancing healthcare research and safeguarding patient privacy in the digital age
Artificial Intelligence Applications and Innovations: AIAI 2024 IFIP WG 12.5 International Workshops, MHDW 2024, 5G-PINE 2024, and AI4GD 2024, Corfu, Greece, June 27–30, 2024 Proceedings
No full textInternational audienceBook Front Matter of AICT 71
Beyond 5G Networking: The Case of NANCY Project
No full textPart 1: The 9th Workshop on “5G – Putting Intelligence to the Network Edge” (5G-PINE)International audienceWith the global deployment of the fifth-generation (5G) wireless networks, the exploration of Beyond 5G (B5G)/sixth-generation (6G) wireless communications has begun. It is foreseen that related technologies will exhibit superior characteristics compared to their predecessors, encompassing higher transfer speeds, expanded coverage, enhanced reliability, greater energy efficiency, reduced latency and, notably, an integrated “human-centric” network infrastructure driven by Artificial Intelligence (AI). This study underscores the imperative need for AI methodologies and security across various resources such as spectrum, computing and storage, provided by the advanced blockchain features in the forthcoming 6G era. Use cases posed by these B5G technologies aim to leverage its inherent features of decentralization, transparency, anonymity and resiliency, while blockchain can foster cooperative trust among disparate network entities. Furthermore, the paper elucidates insights into Blockchain Radio Access Networks (B-RANs) gleaned from the EU-funded research project NANCY [1], whose pillars and architecture are highlighted, providing an overview of the core advancements it can offer