Articles tagged with Computer Modeling
A study found that GPT-4 matched the performance of radiologists in detecting errors in radiology reports, with an accuracy rate of 82.7%. The use of GPT-4 resulted in lower mean correction cost per report than the most cost-efficient radiologist.
A novel machine learning model has been developed to characterize material surfaces, accurately predicting key electronic properties. The model, which employs artificial neural networks and transfer learning, shows great promise for exploring new materials with superior properties.
Mirta Galesic has been awarded a €3 million ERC Advanced Grant to investigate collective adaptation. The project, hosted at the Complexity Science Hub, aims to understand how human groups navigate complex problem landscapes using computational models and empirical data.
Researchers created a system called Holodeck to generate interactive 3D environments, leveraging language models like ChatGPT to control it. The system outperformed earlier tools in evaluating realism and accuracy, with human evaluators preferring its outputs across various indoor environments.
A new machine-learning approach outperforms human testers in generating diverse prompts that trigger a wider range of undesirable responses from chatbots. The technique provides a faster and more effective way to ensure the safety of large language models.
Chemists develop new reactions using model systems and substrates to demonstrate versatility. A new computer-aided method reduces subjective bias by analyzing real pharmaceutical compounds' complexity and structural properties. This improves data quality and facilitates machine learning applications.
Researchers developed a new model to predict future risk of pressure injuries, increasing accuracy by over 20%. The predictive algorithm offers improved economic efficiency and substantial savings, especially for underrepresented minorities.
A team of researchers developed an AI-powered computer vision model to detect Brazilian wild animals on roads and warn drivers in real-time. The system uses roadside cameras and portable computers to identify species such as anteaters, wolves, and tapirs, with the potential to save lives and reduce roadkill.
Researchers at Nagoya University developed a framework for evolving AI agents with diverse personality traits using large-scale language models. The study found that AI agents can switch between selfish and cooperative behaviors, mirroring human behavior.
A new study using machine learning reveals that great gray owls congregate near human-made structures, contrary to previous assumptions. The research used artificial intelligence modeling and large datasets to identify suitable habitats for the owl, providing a more accurate representation of its distribution.
Researchers analyzed over 3,500 river basins worldwide, finding that precipitation was the sole determining factor in only 25% of flood events. Soil moisture and air temperature were decisive factors in around 10% and 3% of cases, respectively. The study suggests that more extreme floods are caused by multiple factors interacting.
Belgian scientists developed AI models to predict beer ratings and aroma compounds, improving beer quality. The study analyzed hundreds of beers and used machine learning to connect chemical concentrations with appreciation scores.
Researchers developed a real-time temperature reconstruction technique for HIFU treatment, enabling accurate monitoring and planning. This approach uses deep learning to transform ultrasonic images into temperature images in just a few milliseconds.
A WVU study found that treating work like a game improves workers' productivity and engagement, but also increases stress levels. Gamification can boost completion times for repetitive tasks, but may push employees past their tolerance threshold, leading to frustration and physical demands.
Researchers at MIT have developed a method to analyze the behavior of granular materials, revealing their internal forces and shapes in 3D detail. This breakthrough may lead to better understanding of landslides and industrial processes.
A recent study investigated ChatGPT-3.5's ability to produce high-quality summaries of medical research abstracts, finding that it was accurate but not always fact-based and prone to minor inaccuracies. The model showed promise as a screening tool to help clinicians quickly evaluate article relevance but should not be relied upon for c...
Researchers have proposed an innovative quantum algorithm that effectively solves combinatorial optimization problems with constraints in a short time. The pVSQA algorithm uses a quantum device to generate a variational quantum state and transform infeasible solutions into feasible ones, achieving near-optimal performance.
Scientists from UC3M and Johns Hopkins University have developed a computational model that simulates the invasion process of cancer cells based on the characteristics of surrounding tissue and cell junctions. The model allows for predicting tumor evolution in patients by analyzing mechanical properties of the microenvironment.
Numerical models help assess ecological impacts of climate change on lakes by simulating changes in phytoplankton composition and community stability. The study found that warming increased seasonal variability in phytoplankton, leading to reduced overall evenness and increased species loss over time.
A new study suggests that removing barriers to cervical cancer screening, surveillance, and diagnostic procedures can save lives and reduce disparities. If every eligible person gets the full screening and follow-up tests recommended, up to 23% fewer people would be diagnosed with cervical cancer and 20% fewer would die from it.
Researchers have developed a new AI model, AsymMirai, to predict 5-year breast cancer risk from mammograms. The model performs almost as well as the state-of-the-art Mirai, but with an interpretable reasoning process, making it a valuable adjunct to human radiologists.
Researchers from UNIGE have developed an AI that can learn a task solely based on verbal instructions, then describe it to another AI, which reproduces the task. This breakthrough is promising for robotics and understanding human language.
Researchers have discovered a universal mechanism for cell motility, applying to various types of migrating cells. Cells tend to move circumferentially on convex structures and prefer axial forward or backward motion on concave surfaces.
A new AI model, called Lars, can accurately detect signs of lymph node cancer in 90% of cases, reducing workload for radiologists and increasing access to healthcare. The model was developed using a large dataset of over 17,000 images from 5,000 patients and is based on deep learning technology.
Researchers developed a computer model that accurately diagnoses pulmonary hypertension in newborn infants using ultrasound images, suggesting correct diagnosis in 80-90% of cases. The model's success relies on highlighting key characteristics, allowing doctors to understand the criteria used for decision-making.
Communications of the ACM has been revamped as a web-first, open-access publication, enabling rapid article publication and enhanced reader experience. The new model aims to increase engagement with the broader computer science community, expand author reach, and showcase cutting-edge ideas and research.
A new traffic signal concept, known as the 'white phase,' uses autonomous vehicles to expedite traffic flow at intersections. The concept has been shown to improve travel time for both pedestrians and vehicles, especially when autonomous vehicles make up a higher percentage of traffic.
Researchers are developing a new framework to integrate renewable energy sources with the power grid using machine learning. The goal is to ensure efficient and stable operations while maximizing wind and solar power usage.
A new study led by Worcester Polytechnic Institute aims to determine whether AI can help doctors predict which patients will benefit from mindfulness-based stress reduction in managing chronic lower back pain. The research uses machine learning and physiological data from fitness sensors to detect patterns that may not be apparent to d...
New research combines images with computer-enabled analysis to tackle biological questions globally. Imageomics aims to improve image classification and analysis using machine learning and computer vision, enabling faster scientific discoveries.
Researchers at Tohoku University have developed a high-performance spin wave reservoir computing model that utilizes spintronics technology. The breakthrough could lead to energy-efficient, nanoscale computing with unparalleled computational power.
A computational model of the more than 26 million atoms in a DNA-packed viral capsid has expanded our understanding of virus structure and DNA dynamics. The study found that the DNA formed switchback loops as it was pushed into the capsid, similar to how DNA is organized in eukaryotic cells.
A new study led by Cleveland Clinic researchers suggests that sildenafil may help protect brain cells from Alzheimer's disease. The study found that sildenafil can lower levels of neurotoxic tau proteins and improve brain function in patients with Alzheimer's.
Researchers argue that current automated toxicity detection methods can improve but require human intervention to review decisions. Companies can take steps to improve working conditions and platform cultures that prioritize kindness and respect.
A team led by Prof. Wolf Gero Schmidt used Hawk supercomputer to study how strategic impurities in solar cells can improve performance. They discovered that certain defects can improve exciton transfer, leading to more energy captured. This breakthrough could lead to more efficient and climate-friendly energy production.
Grant Shields, a psychologist at the University of Arkansas, has received an NSF CAREER Award to investigate how stress affects cognitive control. His research aims to identify the underlying processes that influence inhibitory control and develop a publicly available course to disseminate his findings.
PandaOmics uses advanced AI algorithms to process vast quantities of diverse data, performing gene and pathway analysis and target predictions. The platform has been extensively validated in multiple therapeutic areas, including oncology, inflammation, and immunology.
A recent study by Singapore Management University researchers used machine learning to develop a personalized decision support tool for ICU patients. The analysis indicates that incorporating predictive information can reduce ICU length of stay and extubation failure rates, leading to improved patient outcomes.
A new study from Vanderbilt University Medical Center demonstrates the potential of AI to help target computerized alerts intended for healthcare teams. The approach uses machine learning to analyze user interactions with alerts, predicting when specific alerts will be dismissed by users. The model's suggestions were found to align wit...
Researchers at Hokkaido University developed new computer-based models of masting to understand its effects on ecosystems and food webs. The models predict that masting can affect tree mortality, population dynamics, and animal-forest interactions, with implications for mitigating negative impacts of climate change.
Researchers have introduced an optimization technique that accelerates Bayesian inference without requiring extensive user effort. This new automated method achieves more accurate results faster than another popular approach and offers reliable uncertainty estimates to help scientists understand when to trust their predictions.
Gossiping provides a social benefit by disseminating information about people's reputations, helping recipients connect with cooperative individuals while avoiding selfish ones. This allows gossipers to gain an evolutionary advantage as their actions influence others' behavior and encourage cooperation.
A new computer model of a human lung is being used to simulate the interaction of radiation with lung tissue at a cellular level. This could lead to more targeted treatments for cancer and reduce damage caused by radiotherapy.
Researchers found that widely used machine learning tools produce biased results for immunotherapy research, as they rely heavily on datasets from higher-income communities. This can lead to ineffective treatments for lower-income populations. The study highlights the need for accurate and unbiased data in machine learning models.
A new AI tool, DeepGO-SE, successfully predicts the molecular functions of unknown proteins with high accuracy. This breakthrough enables researchers to analyze uncharacterized proteins, facilitating tasks such as drug discovery, metabolic pathway analysis, and disease associations.
Researchers from Argonne National Laboratory and the University of Illinois Urbana-Champaign used generative AI to quickly assemble over 120,000 new MOF candidates for carbon capture. The approach combines AI with high-throughput screening, molecular dynamics simulations and theory-based design to identify optimal materials.
Researchers use machine learning to combine mismatched datasets and reduce variation by over 95%, retaining meaningful differences. The approach has potential to provide deeper understanding of normal metabolism and identify biomarkers for disease.
Researchers at UBC Okanagan have developed a new modeling framework called TOSCA, which helps improve wind energy forecasts and productivity. The framework can capture the interaction between large wind farms and the oncoming wind, leading to more accurate estimates of power output.
Researchers use advanced electron microscopy and computational modeling to understand tantalum oxide formation, which can impede qubit performance. The study reveals a 'suboxide' layer at the interface between tantalum and oxide, with ordered crystalline lattice features.
Researchers at George Mason University are exploring security concerns around prompts for large language models (LLMs), focusing on attack and defense strategies with explainable AI and human involvement.
Researchers at University of Virginia Health System developed a new approach to machine learning that identifies drugs minimizing harmful scarring after heart attacks. The tool predicts and explains drug effects for other diseases as well.
Researchers created an AI-enabled model to help mitigate global ammonia emissions from agriculture. By optimizing fertilizer management, the model can effectively reduce emissions by up to 38%, with Asia having the highest potential for reduction.
A multidisciplinary team created a new algorithm to provide optimal personalized treatment options for individuals with esophageal cancer. The framework uses various heterogeneous factors that facilitate the progression of the disease and demonstrated high accuracy in controlling esophageal cancer.
Researchers developed a novel AI technique to detect Medicare fraud in big data, using Random Undersampling and supervised feature selection. The method improved classification performance, especially when reducing features, and outperformed models with all available features.
Physicists at Leipzig University have developed a neural network that uses active colloidal particles for artificial intelligence. The system reduces noise and increases efficiency in calculations by utilizing past states of the reservoir.
A new study by Northwestern University found that lab-trained AI models are easily misled by tissue contaminants, resulting in errors in diagnoses and vessel damage detection. The researchers suggest improving the problem of quantifying and addressing biological impurities in AI models to enhance accuracy.
Researchers found that more extreme and frequent droughts could lead to a nearly five-fold increase in total area burned, from 231 sq mi in 2016 to over 17,000 sq mi by 2100. Fires would also become more frequent, returning to certain forest points every five years.
Researchers developed an AI-driven algorithm that accurately predicts death and complications after percutaneous coronary intervention (PCI). The tool uses patient feedback to provide a patient-centered, individualized risk prediction score.
Researchers at ISTA investigated the crucial set of synapses between neurons within the cerebellum, uncovering details of their function and development. The study used advanced techniques to look at the inhibitory synapses in great detail, revealing how they delicately influence the cell's signal output.
Researchers analyze tidal disruption events (TDEs) to estimate the properties of supermassive black holes and stars. The CN22 model, proposed by Syracuse University researchers, provides a new way forward for understanding TDEs and their implications for galaxy evolution.