Articles tagged with Artificial Intelligence
Researchers created an artificial sensory receptor that generates spike signals on its own, enabling the e-skin to analyze spatial information and react to external stimuli in real-time. The e-skin's functionality overcomes limitations of conventional electronic skins, which can only process tactile information sequentially.
Researchers analyzed 60 billion tweets to understand vaccine hesitancy on social media. They found that anti-vaxxer profiles often share links to YouTube videos and commercial sites selling alternative health products, highlighting the spread of misinformation in echo chambers.
Developed by University of Seville researchers, the new methodology has a sensitivity of 100% and specificity of 87.5%. It can detect SARS-CoV-2 in saliva and synthetic viruses with minimal equipment and training.
Researchers used AI to identify detrimental dams and reveal lost environmental benefits from existing 158 hydropower dams. The study considers six socio-environmental criteria, including river flow and energy production, to optimize dam selection across the entire Amazon basin.
Researchers analyzed new kinds of atomic-scale microscopic images using artificial intelligence to understand why rechargeable batteries wear out. They discovered nanofractures caused by mechanical strain on materials, which could lead to the development of more indestructible batteries.
The American Roentgen Ray Society recognizes Drs. Nguyen, Beheshtian, Harfouch, and Hashiba for their outstanding research and education contributions to the field of radiology. The awards honor the candidates' scientific merit and potential impact on imaging and allied sciences research.
A machine-learning algorithm named MAD3 can predict mechanical properties of metals without performing physical tests, cutting testing time by 1,000 times. The algorithm replaces traditional simulation software, enabling faster research and development with minimal resources.
Researchers have developed an AI-based method to analyze cryo-electron microscopy data, enabling the simultaneous examination of multiple protein complexes in cells. This breakthrough can lead to a better understanding of protein functions and potentially create new treatments for diseases like Alzheimer's and cancer.
Researchers found AI-synthesized faces to be nearly indistinguishable from real faces and rated as more trustworthy. The study's results have significant implications for the spread of manipulated images, including potential use in revenge porn and propaganda.
A study published in The Journal of Finance and Data Science found that using machine learning to estimate reward and risk improves portfolio performance. Machine learning helps provide accurate estimates by taking advantage of complex non-linear relationships between market variables, outperforming linear methods.
Researchers from Argonne National Laboratory have created a set of new practices to guide the curation of high energy physics datasets, making them more FAIR and reusable. The goal is to automate the finding and use of data for humans and streamline the development of AI tools for scientific discovery.
Researchers developed a new method using smartphone vibration motor and camera to monitor blood clots, with accuracy similar to commercially available techniques. The device can be used at home, making it easier for patients to test their PT/INR levels, potentially reducing the need for frequent clinical visits.
Researchers developed an AI system called GestaltMatcher that uses facial characteristics to detect rare diseases with high accuracy. The system considers similarities between patients and can even suggest diagnoses for previously unknown diseases.
Researchers found that consumers respond better to AI agents when a product offer is worse than expected, but more favorably to human agents when an offer is better than expected. Designing AI agents to appear more humanlike can change consumer response.
A team of scientists has developed a pioneering approach to combine advances in computer vision with ecological expertise to analyze wildlife populations. By leveraging AI and machine learning algorithms, researchers can extract key features from images and videos to quickly classify species, count individuals, and track behavior.
The grant aims to create toolkits and training for AI developers to prevent existing structural inequalities from becoming embedded into emerging technology. Researchers will investigate the impact of AI on core human values, including democratic rights and minority rights.
Researchers at Purdue University have created a device that can dynamically rewire itself to adapt to new data, enabling artificial intelligence to learn and remember information like the human brain. This breakthrough could lead to more efficient AI systems for tasks such as image recognition and decision-making.
Researchers developed an algorithm that incorporates customer behavior into recommendation systems, making more accurate and personalized suggestions. The technique, known as tensor decomposition, analyzes data in multiple dimensions to capture complex patterns and relationships.
Researchers developed a machine learning algorithm to automate propofol dosing for unconscious patients, matching human performance in sophisticated simulations. The 'dose penalty' model improved upon traditional software, but limitations remain, highlighting challenges in AI system accuracy and real-world application.
A new technique uses compression to reduce data transmission size, allowing for efficient federated learning on wireless devices. The approach has been shown to condense data packets by up to 99%, making it suitable for areas with limited bandwidth.
A team of researchers from the Institute of Industrial Science, The University of Tokyo, used a mathematical model to examine the implications of intergenerational learning. They found that learning accelerated the evolutionary process, which may assist in designing more efficient hybrid algorithms.
Researchers from UOC-led OptimalSharing@SmartCities project will analyze inhabitants' mobility patterns and demands to design more efficient shared transport practices. The project aims to develop agile optimization algorithms capable of processing large volumes of data in real-time for dynamic system coordination.
A new AI-powered system tracks food intake to ensure residents meet their nutritional needs, providing accurate data and reducing errors. The system analyzes photos of plates after meals, estimating food consumption and calculating its nutritional value.
Researchers at OHSU are developing a new approach to scientific imaging to study the dynamic organization of cells by examining how tiny molecules make cells work. The W. M. Keck Foundation has awarded $1 million to develop a one-of-a-kind imaging and computational system.
Researchers developed MonoCon, a new AI technique that enables accurate identification of 3D objects in 2D images. By incorporating auxiliary context, the method improves object detection and estimation accuracy, paving the way for safer and more robust autonomous vehicles.
Researchers at the University of Groningen have developed an AI system that can recognize indoor spaces with high accuracy by combining image and audio data. The system achieved a 70% accuracy rate in recognizing nine different types of indoor spaces, surpassing previous results.
A robot performed laparoscopic surgery on a pig without human guidance, demonstrating improved results compared to human surgeons. The Smart Tissue Autonomous Robot (STAR) excelled at intestinal anastomosis, a procedure requiring high precision and accuracy.
Researchers from KTU proposed a deep-learning-based method for 3D human shape reconstruction using limited-angle depth data. The method can be integrated with existing virtual reality tools and has potential applications in telemedicine and remote diagnostics.
A team of scientists developed an AI-based model to predict personal thermal comfort based on spatial parameters, achieving exceptional accuracy. The study highlights the importance of incorporating architectural features in models to reduce energy consumption.
A new software uses pose estimation to track human motion with high accuracy, providing an objective assessment of motor function. The technology has the potential to revolutionize neurological care by enabling patients to record video that can be analyzed by their physicians remotely.
Physicists have detected X particles in quark-gluon plasma produced in the Large Hadron Collider, a phenomenon that could reveal the particles' unknown structure. The discovery uses machine-learning techniques to sift through massive datasets and identify decay patterns characteristic of X particles.
A team of researchers has successfully developed a material-based reservoir computing device using single-walled carbon nanotubes and porphyrin-polyoxometalate composites. This innovation enables AI robots to classify grasped objects based on tactile signals, demonstrating the potential for energy-saving AI systems and situational awar...
A joint research team at KAIST developed a technique for facial expression detection using near-infrared light-field cameras and AI technology. The resulting NIR-based light-field camera (NIR-LFC) acquired high-quality 3D reconstruction images of facial expressions regardless of lighting conditions.
MIT researchers develop teaching phase that guides humans in understanding AI strengths and weaknesses, enabling more accurate decisions and faster conclusions. The technique helps humans build a mental model of the AI agent, reducing reliance on biased assumptions.
Researchers at Brigham and Women's Hospital conducted a scoping review of 78 articles to identify key use cases for AI in preventing or mitigating adverse drug events (ADEs). Genetic information is thought to be critical in improving AI algorithm performance, but systematic evaluations are necessary to generate evidence for this field.
MIT researchers develop a method to test feature-attribution methods for machine-learning models. They find that even the most popular methods often miss important features in an image and some perform as poorly as a random baseline. This has major implications for high-stakes situations like medical diagnoses.
A study by the University of Texas at Austin found that software development teams given greater autonomy are more productive and have higher customer satisfaction rates. The researchers tested 461 projects over 50 months and found a 39% increase in value added for autonomous teams compared to traditional teams.
A team of scientists from Gwangju Institute of Science and Technology developed a deep learning-based approach to predict SC2 battle outcomes by considering army composition and terrain type. The proposed model leveraged parameter sharing, enabling it to analyze complex factors accurately and make predictions.
A recent study by MIT researchers found that family members trust devices with human-like social behaviors, such as Amazon's Alexa or Jibo's social robot. The study revealed significant effects of branding on perceptions, showing users viewed Google as more trustworthy than Amazon despite similar designs and functionality.
Researchers at RIKEN CBS demonstrate that neural networks minimize energy cost and solve mazes efficiently, pointing to a set of universal mathematical rules. The findings will aid in analyzing impaired brain function and generating optimized neural networks for artificial intelligences.
A new algorithm, FusionM4Net, has been developed to classify skin lesions with improved diagnostic accuracy. The algorithm uses a multi-stage data fusion process and outperforms previous state-of-the-art algorithms.
Researchers at Duke University developed an artificial intelligence platform that analyzes potentially cancerous lesions in mammography scans. The algorithm is interpretable, showing physicians how it came to its conclusions, making it more trustworthy and useful for training students and aiding physicians in sparsely populated regions.
A large international evaluation shows AI systems can identify and grade prostate cancer in tissue samples from different countries equally well as human pathologists. The results suggest AI can be a complementary tool in prostate cancer care, improving diagnostic quality and consistency.
An AI-driven solution identifies sites on RNA and DNA molecules where interaction with potential drug candidates can occur. This allows pharmaceutical companies to discover new medications in a more focused and efficient manner.
Researchers found that two commonly used atomic fingerprints, ACSF and SOAP, are insensitive to certain movements, leading to the failure of machine learning in resolving four-body interactions. This limitation affects the accuracy of reproducing these interactions with limited success.
Scientists have developed a mathematical method to interpret data on underground water flows, providing more efficient and accurate imaging for planning construction works and inspecting dams. The technique has great potential for locating water reservoirs in dry areas and tapping into this resource for agricultural and industrial needs.
The European Research Council has awarded €1.5 million in grants to three Saarbrücken-based researchers. Their projects focus on artificial intelligence and cybersecurity, with aims to develop fairer machine learning algorithms and secure computing methods.
A team of researchers from the University of Tokyo found that public trust in AI varies greatly depending on the application and demographic factors. They developed an octagonal visual metric to quantify these attitudes and hope it can lead to a universal scale for measuring ethical issues around AI.
A new machine learning algorithm can accurately identify fish calls from hydrophone data, enabling efficient analysis of marine environments. The method has implications for understanding changes in ocean conditions and coral reef health.
Researchers found that three algorithms - Multilayer Perceptron, Fuzzy Cognitive Map, and Deep Neural Network - outperformed others in diagnosing COVID-19 at early stages. These findings can guide software development to create intelligence-based tools for early diagnosis.
A study found that consumers trust companies more when interacting with chatbots that exhibit high perceived humanness, whether powered by AI or humans. This suggests that companies should focus on creating conversational interfaces that convey a sense of investment and relationship, rather than just relying on human agents.
A pilot study suggests that machine learning algorithms combining EKG and electronic health record data can more effectively screen for pulmonary embolisms than current tests. The fusion model was found to be 15-30% more effective at accurately identifying cases, especially severe ones.
A new neural net technique combines optical imaging and synthetic-aperture radar to provide a single comprehensive data source, enhancing disaster response capabilities. The approach achieves up to a 96% accuracy rate in detecting objects and can be applied to various tasks such as crop health monitoring and building identification.
The NYU Tandon team is developing an AI 'virtual assistant' to provide just-in-time visual and audio feedback for complex tasks. The project aims to create a dynamic intelligent agent that engages with users, responding to their emotional state, location, and surrounding conditions.
A new study has generated a massive dataset of artificial voice recordings, revealing subtle differences in high frequencies between real and fake files. The researchers developed algorithms that can distinguish between deepfakes and real speech, providing a starting point for other researchers to improve detection methods.
The Penn Artificial Intelligence and Technology Collaboratory for Healthy Aging (PennAITech) aims to explore the use of artificial intelligence to improve in-home care for older adults and individuals with Alzheimer’s disease. The collaboratory will leverage expertise in AI, consumer informatics, and aging from various departments acro...
Researchers analyzed hundreds of thousands of secure email messages between doctors and patients to find that most doctors use language too complex for their patients' low health literacy. Effective communication can improve patient outcomes by tailoring electronic messages to match the complexity of the patient's language.
Bielefeld University researcher Yaochu Jin is working on developing artificial intelligence that mimics nature's principles to tackle complex problems. His research aims to create more flexible AI systems that can adapt to uncertainty and fuzziness, with potential applications in healthcare and other fields.
A new study published in JAMA Network Open found that artificial intelligence can predict which people attending memory clinics will develop dementia within two years with 92% accuracy. The machine learning algorithm analyzed data from over 15,300 patients and identified around eight per cent of dementia diagnoses as being made in error.
Researchers developed an AI model that can diagnose COVID-19 with high accuracy, using federated learning to preserve patient data privacy. The model was trained on over 9,000 CT scans from 23 hospitals in the UK and China, and validated against a panel of radiologists.