Articles tagged with Artificial Intelligence
The RITHMS project aims to enhance police forces' and customs authorities' capabilities in tackling illicit cultural trafficking with innovative technologies. The platform will utilize Social Network Analysis to identify organized criminal networks and provide valuable information to investigators.
A digital detection of dementia study aims to evaluate the practical use of an AI tool for early identification of Alzheimer's disease and related dementias. The study uses a passive digital marker and patient-reported outcomes survey, which have shown potential in detecting mild cognitive impairment and reducing healthcare costs.
Researchers created an AI platform called DystoniaBoTXNet that uses brain MRIs to identify patients who will respond to botulinum toxin treatment. The platform achieved a 96.3% accuracy rate in predicting treatment efficacy, offering clinicians a new tool for refining clinical decisions.
A recent study found that AI failed to pass a radiology qualifying examination, with an average accuracy of 79.5%. However, the researchers suggest that further training and revision could lead to improved results.
Researchers at Georgia Institute of Technology found that AI chatbots' positive emotional displays have no effect on customer service evaluations when the source is a bot, but improve when human agents exhibit emotions. The study suggests businesses should be cautious about equipping AI agents with emotion-expressing capabilities.
The UTSA-led Consortium on Nuclear Security Technologies (CONNECT) has received a five-year, $5 million grant from the U.S. Department of Energy's National Nuclear Security Administration. The program aims to educate and train the next generation of scientists and engineers in nuclear security, with a focus on underrepresented students.
Researchers are conducting on-site surveys and generating high-resolution damage maps for 20-square-mile region affected by the Category 4 storm. The goal is to inform protection efforts and help communities recover from the disaster.
Students from the WüSpace association are working on a project to develop intelligent sensors and algorithms for small satellites, aiming to detect dangerous approaches and prevent collisions. The project is funded by the German Space Agency and allows students to gain hands-on experience in space technology development.
The UTEP-led Computing Alliance will receive $4.8M from Google to improve diversity in computer science fields. The project aims to attract, prepare and support Hispanic students for graduate degrees, with initiatives including lab design, financial support and research collaborations.
Researchers developed an AI-based neural network to detect early knee osteoarthritis from x-ray images, matching doctors' diagnoses in 87% of cases. This method could help reduce unnecessary examinations, treatments, and even knee joint replacement surgery.
Researchers have developed a robot capable of sorting, manipulating, and identifying microscopic marine fossils. Forabot uses robotics and artificial intelligence to automate the tedious process of evaluating foram shells and fossils.
Researchers at Incheon National University have developed an IoT-enabled, real-time object detection system for autonomous vehicles. The YOLOv3-based model achieved high accuracy (>96%) in detecting 2D and 3D objects, outperforming other state-of-the-art detection models.
Researchers found that longer genes are linked to longer lifespans and shorter genes to shorter lifespans, with aging causing a shift towards short genes. This imbalance leads to subtle changes in thousands of genes across various tissues, suggesting systems-level changes contribute to aging.
Researchers at MIT have developed a scheme for private information retrieval that is about 30 times faster than other comparable methods. The technique enables users to search an online database without revealing their query to the server, with potential applications in private communication and targeted advertising.
Researchers explored how customers react to positive emotions from AI chatbots compared to human agents. They found that adding emotions doesn't necessarily improve service quality or customer satisfaction.
Researchers have developed a scaled-up version of a probabilistic computer using stochastic spintronic devices, suitable for combinatorial optimization and machine learning. The new design combines conventional semiconductor chips with modified spintronic devices, achieving massive improvements in throughput and power consumption.
Researchers at FAU and industry partners aim to create a universal radio adapter for seamless and secure operations through non-cooperative indigenous 5G networks. The project aims to reduce interception, disruption, or jamming of communications over 5G networks.
A massive collaborative study using federated learning developed a model that enhances identification and prediction of boundaries in three tumor sub-compartments without compromising patient privacy. The dataset, comprising 6,314 glioblastoma patients from 71 sites globally, is the largest and most diverse ever considered.
A team of researchers from the University of Pennsylvania has developed a new algorithm, metadynamics, that can navigate high-dimensional energy landscapes to find low-energy configurations. This breakthrough has the potential to revolutionize fields such as protein folding and machine learning.
A new AI evaluation framework, GOPHER, has been developed to assess the efficiency of genome analysis algorithms. The tool judges programs on their ability to learn genomic biology, predict patterns, handle noise, and provide interpretable decisions.
Researchers at University of South Australia have developed a non-contact system to accurately measure systolic and diastolic blood pressure using AI algorithms and a camera. The system achieved 90% accuracy in readings, outperforming existing methods.
A study led by Kyoto University researchers found that AI-generated haiku poems, created without human intervention, were often indistinguishable from those penned by humans. In contrast, human-AI collaboration produced more creative works.
A team of researchers developed a neural network that can analyze digital biopsy images to diagnose breast cancer, identifying the presence of PD-L1 protein. The AI system outperformed human pathologists in 70% of cases, with high accuracy rates and potential for personalized medicine.
Researchers used AI to design and test thousands of functional group patterns on a carbon nanotube pore, finding optimal arrangements that can filter out contaminants. The study demonstrates AI's potential in developing new types of water purification membranes.
A neural network trained using a diverse dataset outperforms conventionally trained algorithms by reducing bias in artificial intelligence. The use of images from low-resource populations boosts the object recognition performance of machine learning systems.
Researchers from Kaunas University of Technology and Lithuanian University of Health Sciences developed an AI-based system to evaluate the 'substitute voice' of laryngectomy patients, allowing for faster detection of pathologies. The Acoustic Substitution Voice Index (ASVI) uses acoustic parameters and artificial intelligence methods t...
Researchers propose six competency domains for AI use in primary care: foundational knowledge, critical appraisal, medical decision making, technical use, patient communication, and awareness of unintended consequences. These competencies aim to harness AI's potential for patient and societal benefit.
Concordia researchers identify vulnerabilities in smart inverters to cyberattacks, including reconnaissance, replay, DDoS, and Man-in-the-Middle attacks. These attacks can disrupt energy flow and create power oscillations, severely impeding microgrid functionality.
The researchers have developed an AI algorithm called M3GNet that can predict the structure and dynamic properties of any material. The algorithm was used to create a database of over 31 million yet-to-be-synthesized materials with predicted properties, facilitating the discovery of new technological materials.
Researchers at the University of Texas at Dallas have developed a computer-based platform for drug discovery using topological data analysis. The approach allows for virtual screening of thousands of compound candidates, narrowing them down to the most promising ones for laboratory and clinical testing.
A new study using artificial intelligence has found that a simple eye test can accurately predict the risk of heart disease. The researchers developed an algorithm that can analyze retinal images to assess cardiovascular health, providing a non-invasive alternative to traditional risk scores.
A new intelligent observer for Esports has been developed using a Mask R-CNN algorithm to learn from human spectating data. This approach overcomes limitations of existing automatic observers, which require extensive domain knowledge and cannot capture undefined events or changes in event significance.
Researchers developed a new machine-learning framework that enables cooperative or competitive AI agents to consider the future behaviors of all agents, not just their teammates or competitors. This framework, FURTHER, uses two modules: an inference module and a reinforcement learning module, to enable agents to adapt their behaviors a...
Chugh argues that AI risk assessments raise concerns about human biases and the absence of individualized decision-making in the judiciary. She advocates for more research on AI's role in the court system, prioritizing community-driven and individualized processes.
Chung-Ang University researchers propose a new algorithm, MR-UCB and MR-APE, to tackle stochastic multi-armed bandit problems with heavy-tailed noise distributions. The methods guarantee minimal loss for worst-case scenarios with minimal prior information.
A research team developed an optical chip that can train machine learning hardware, improving AI performance and reducing energy consumption. This innovation uses photonic tensor cores and electronic-photonic application-specific integrated circuits to speed up the training step in machine learning systems.
A new research project at Aarhus University aims to develop intelligent drones that can detect ice on turbine blades, optimizing energy production and expanding market opportunities. The project has the potential to reduce energy losses by up to 80% and enable wind farms to operate in colder climates.
A new AI tool helps governments decide whether to bail out a bank by predicting if the intervention will save money for taxpayers. The algorithm assesses financial implications and suggests optimal bailout strategies.
Researchers at Singapore University of Technology and Design (SUTD) have developed a novel phase-change key for new hardware security. The device, known as the physical unclonable function (PUF), is scalable, energy-efficient, and secure against AI attacks compared to traditional silicon PUFs.
Researchers used weather radar to track bird movements and found peak roosting stages shifting earlier due to warmer temperatures. This shift may lead to a shortened pre-migratory season, impacting birds' survival during migration.
Researchers at Oak Ridge National Laboratory have discovered genetic markers for autism, developed recyclable composites to drive the net-zero goal, and created a tool for real-time building evaluation. Additionally, they have made significant progress in growing hydrogen-storage crystals using a novel nano-reactor material.
Researchers have developed wearable electronics paired with artificial intelligence to detect emerging health problems, such as heart disease and cancer, before symptoms appear. The device can perform personalized analysis of tracked health data while minimizing wireless transmission.
Researchers created a Facebook-like prototype platform, called Trustnet, where users rate posts as accurate or inaccurate before sharing. This approach showed that people effectively assess misinforming posts and share their assessments with others.
Researchers at Mayo Clinic developed an AI algorithm that accurately detects weak heart pumps using single-lead Apple Watch ECGs. The system has shown promising results, rivaling medical treadmill diagnostic tests in accuracy, and holds potential for scalable screening and prevention of heart failure.
Researchers developed an AI tool that analyzes plaque features on coronary CT scans to predict reduced blood flow, potentially reducing invasive tests. The tool could help doctors determine the next step in treatment plans and risk-stratify patients correctly.
A team of researchers used AI pattern recognition to re-analyze footprints from the Dinosaur Stampede National Monument and concluded that they were made by an ornithopod dinosaur, a herbivorous species. The results contradicted the long-held assumption of a vicious dinosaur predator.
The study developed an extended deep Q-network (EDQN)-incorporated context-based meta-RL model that can autonomously detect traffic states, classify regimes, and assign signal phases. The model outperformed existing algorithms in simulation experiments and showed adaptability to new tasks without adjusting parameters.
Researchers created a new set of standards, called FAIR, to manage AI models, making them findable, accessible, interoperable and reusable. This standardization enables cross-pollination across teams and reduces duplication of effort, ultimately facilitating scientific discovery.
Experts warn that poorly functioning IT systems are a clear and present threat to patient safety, resulting from inadequate investment and lack of prioritization. The British Medical Association estimates that 27% of clinicians lose over four hours a week due to inefficient IT systems, highlighting the need for urgent improvement.
The DUNE project uses deep learning techniques and distributed computing to improve indoor positioning accuracy. The system combines various sensor technologies and adapts to different industrial needs, providing real-time updates and reducing margins of error.
Researchers have demonstrated a power-efficient component for demultiplexing operation using silicon photonic MEMS, enabling efficient wavelength demultiplexing for fiber-optic communications. The compact footprint of the add-drop filter allows fast operation compared to established MEMS products.
The University of Tsukuba researchers developed a machine-learning model to predict undertriage in phone-based triage systems. The model identified risk factors such as age, sex, and comorbid conditions, which can be used to update protocols and improve patient outcomes.
A new AI technology has been developed to automatically diagnose lung diseases such as tuberculosis and pneumonia, freeing up radiographers and reducing waiting times for test results. The system, which uses X-ray imaging and deep learning algorithms, achieved a 98% accuracy rate during extensive testing.
The National Science Foundation has awarded $800,000 to support the development of Edge AI applications in a proposed wearable diabetes management device. The project aims to create a smart, wearable device that monitors blood sugar levels without the need for frequent blood draws, reducing delays and increasing privacy and security.
NJIT Assistant Professor Salam Daher is using augmented reality technology to train caretakers for the elderly. The project aims to improve communication and empathy skills through a digital model of an older person that responds to emotions and environment. The system will be tested to investigate its impact on patient outcomes.
Researchers developed an intelligent compaction technology that integrates into a road roller, assessing real-time the quality of road base compaction. This improves road construction, reducing potholes and maintenance costs, leading to safer and more resilient roads.
A new research group aims to develop an interdisciplinary approach for secure operation of critical infrastructure systems using artificial intelligence. The goal is to create a 'hybrid architecture' that combines AI-based strategic development with rule-based systems, ensuring safe and reliable behavior.
Researchers at the University of Houston are developing a clinical decision support system to predict which patients with diabetes are most likely to experience complications. The tool, called Primary Care Forecast, uses deep learning to consider patient health history and social factors such as employment status and education level.
Researchers from Osaka University have developed an AI-powered method to identify optimal amino acid mutations in enzymes. This approach accelerates the enzyme engineering process, allowing for tailored enzyme designs suitable for various biochemical environments.
Deep learning models can become less accurate in recognizing specific categories of images, sounds, or text after network pruning. Researchers demonstrate a technique to address this challenge, improving the fairness of deep learning models.