Articles tagged with Computer Modeling
A novel robotic system developed by USC researchers can help clinicians accurately assess a patient's rehabilitation progress. The method generates an 'arm nonuse' metric using machine learning and a socially assistive robot to track how much a patient is using their weaker arm spontaneously.
Researchers at Oak Ridge National Laboratory used quantum biology and artificial intelligence to sharpen the CRISPR Cas9 genome editing tool, improving its efficiency on microbes. The new model revealed key features about nucleotides that enable better guide RNA selection.
Researchers at UC Berkeley introduce prediction-powered inference (PPI), a method to correct machine learning model output and provide valid confidence intervals. PPI allows scientists to incorporate AI predictions into their work without making assumptions about the model's limitations or data biases.
The UTSA MATRIX AI Consortium has received a $2 million grant to create new AI models that rapidly learn, adapt, and operate in uncertain conditions. The team aims to bridge the gap between human brain processing efficiency and current AI limitations, enabling more efficient and adaptive AI systems.
Yu Yang's NSF-funded research aims to reduce vehicle emissions and promote the use of electric bikes and scooters by developing socially informed traffic signal control systems. The project involves a three-pronged method that uses low-cost mobile air-quality sensing, spatial-temporal graph diffusion learning, and reinforcement learnin...
Researchers used computer modeling to analyze cancer incidence and mortality rates among diverse populations, confirming the validity of the approach. The studies aimed to address critical public health issues by identifying leverage points to increase equity in cancer burden among Black populations.
Researchers redesigned IVF needles to reduce fluid flow damage to eggs, improving oocyte collection and IVF success rates. The study uses computer models and mathematical simulations to optimize the design, which has been successfully tested in cattle, with plans for human trials.
Researchers developed a deep learning system to detect and predict joint space narrowing and erosions in hand radiographs of RA patients. The AI model achieved over 90% accuracy in detecting joints, but its performance may be lower than human radiologists for wrist joint analysis.
A UNIGE team has developed a super-model to simulate the spread of three green technologies in Swiss municipalities by 2050. The results show that Switzerland is unlikely to achieve zero net carbon emissions by 2050 without significant policy changes, highlighting the need for increased efforts and updated policies.
Research from Harvard John A. Paulson School of Engineering and Applied Sciences estimates that humans have increased atmospheric mercury levels sevenfold, with a pre-anthropogenic baseline of around 580 megagrams. Human emissions from coal-fired power plants and waste-incineration are responsible for the majority of this increase.
Researchers are combining biology, physics, computer science, and engineering to design electric circuits that mimic the brain's adaptive behavior. The goal is to create a more efficient AI application that can learn from history and adapt without significant energy consumption.
A team of scientists discovered two types of neurons in fruit flies and mice that enable them to identify distinct smells. With experience, these animals can learn to differentiate between very similar odors, a process that could improve machine-learning models and AI systems.
Researchers found self-supervised models generate activity patterns similar to mammalian brains, suggesting an organizing principle. The models learn representations of the physical world to make accurate predictions, potentially unlocking human-labeled data limitations.
Researchers analyzed whiteschist from the Dora Maira Massif to study rapid upward movements, revealing a sharp decrease in pressure or decompression. This suggests that UHP rocks may not have reached a depth of 120 kilometers before returning to the surface.
A new method called TWC-Swin effectively restores holographic images even under low spatial coherence and arbitrary turbulence, surpassing traditional convolutional network-based methods. The study demonstrates strong generalization capabilities, extending its application to unseen scenes.
A new study suggests that Venus once had plate tectonics similar to those on early Earth, which could have supported microbial life. The researchers used atmospheric data and computer modeling to show that the planet's current atmosphere and surface pressure would only be possible with an early form of plate tectonics.
A University of Córdoba team developed an algorithm that predicts student performance in online education, dividing students into four categories and providing personalized assistance. The algorithm uses ordinal classification and fuzzy logic to make more accurate predictions than previous models.
Researchers at MIT found that similarity-focused generative AI models falter when tasked with designing new products, highlighting the need to prioritize innovation in engineering tasks. By adjusting training objectives and metrics, AI can be an effective 'co-pilot' for engineers, enabling faster creation of innovative products.
Researchers developed a novel physical theory that can accurately predict protein folding, surpassing existing models like AlphaFold 2. The new model, WSME-L, can elucidate folding processes without limitations, enabling a comprehensive understanding of protein structures and behaviors.
A recent study discovered two subgroups of people with idiopathic generalized epilepsy, one experiencing highest incidence during sleep and the other during daytime. The researchers found that either dynamics of cortisol or sleep stage transition explained most of the observed distributions of epileptiform discharges.
A new computer model reconstructs the evolution of Alpine ice cover with unprecedented precision, allowing scientists to understand past climate interaction with glaciers. The simulation provides a direct visualization of phenomena, making them accessible to a wide audience.
Research investigates impact of storm fronts, tropical storms, and cyclones on ocean circulation, finding changes in atmospheric synoptic variability (ASV) slow down ocean circulation and decrease primary productivity. ASV variations also affect mixing of ocean's layers and strength of oceanic circulation systems.
A study reveals that variable C:N:P ratios of phytoplankton are essential for regulating dissolved oceanic nutrient ratios, while also influencing atmospheric CO2 levels on geological time scales. The findings challenge the commonly hypothesized strong link between phytoplankton and seawater nutrient ratios.
A team has developed a comprehensive platform called HydroBIM for digital design, intelligent construction, and smart operation of hydropower engineering projects. The platform improves efficiency by 1.5 to 2 times compared to conventional methods, enhancing cooperation, work efficiency, and information integration.
A new model describes microswimmer self-propulsion energy requirements, enabling optimized shape designs and applications in microfluidics, biophysics, and material science. The study reveals surprising similarities between artificial and natural shapes.
Researchers at the University of Virginia Health System are developing computer models to better understand the cellular processes and gene activity of multi-drug resistant bacteria Staphylococcus aureus and Pseudomonas aeruginosa. The goal is to identify vulnerabilities in these bacteria and advance the development of new treatments.
Researchers tested the feasibility of using locally run LLMs like Vicuna-13B to label key findings in chest X-ray reports while preserving patient privacy. The results showed moderate to substantial agreement with non-LLM computer programs, suggesting that these models can be a viable option for AI research.
A new generative model named scPoli enables multi-scale representations of cells and samples, facilitating the integration of high-quality large-scale datasets for novel biological insights and disease understanding. This model accelerates atlas building and usage, ultimately accelerating disease understanding and therapy development.
Post-Acute Sequelae of COVID-19 research aims to track long-term health symptoms in survivors. A $3.7 million grant will support the development of self-supervised deep learning technologies to recognize post-COVID lung progression phenotypes.
Researchers from Imperial College London and the University of Nottingham used machine learning to identify 'atomic shapes' that form basic pieces of geometry in higher dimensions. The findings reveal unexpected patterns in these shapes and demonstrate the potential for machine learning to accelerate mathematical discoveries.
Researchers developed a novel approach to predict therapeutic targets for aging and age-related diseases. They trained a domain-specific BioGPT model on biomedical literature, which improved its performance in identifying prospective targets.
A new AI model integrates imaging and non-imaging patient data for improved diagnostic performance on chest X-rays. The multimodal model outperformed other models for diagnosing up to 25 conditions, showing potential as an aid to clinicians in high-pressure diagnoses.
A machine learning model developed by researchers can predict lung cancer risk with an 83.9% sensitivity, identifying those who need screening for the disease. The model uses just three predictors: age, smoking duration, and pack-years of smoking, making it a simplified approach to determine high-risk populations.
A new AI method leverages causal relationships in genome regulation to efficiently identify optimal genetic perturbations for cellular reprogramming. The technique reduces experimental costs by prioritizing the most informative interventions, leading to faster convergence and more effective results.
A University of Virginia team has developed a new analytical tool using hydrogels to cultivate vascular sprouting from mouse lung tissue, providing new insight into idiopathic pulmonary fibrosis. The research aims to understand the biomechanical and biochemical cues to blood vessels in the lungs during disease progression.
A new modeling method powered by interconnected processors removed human bias from the debate over dinosaurs' demise. The study suggests that the outpouring of climate-altering gases from the Deccan Traps alone could have been sufficient to trigger global extinction, consistent with volcanic eruptions contributing to the mass extinction.
A new clinical and research partnership has created an AI model that can predict whether cancerous tissue has been fully removed from the body during breast cancer surgery. The model performed as well as humans in identifying positive margins, especially in patients with higher breast density.
Scientists developed computational eye models to help patients and surgeons select ideal intraocular lenses and predict visual outcomes. The technology uses anatomical information of the patient's eye to provide guidance on expected optical quality post-operatively.
The project aims to assess the operational resilience of microgrids on DoD installations and ships, using new operational resilience indexes developed by Lehigh University researcher Javad Khazaei. The team will develop a dashboard to monitor resilience indexes in real-time, providing recommendations for improving the systems.
A recent study found that dopamine release makes decisions faster but tends to be less accurate. Researchers analyzed response times to understand the role of dopamine in decision-making processes, revealing a speed-accuracy trade-off.
Researchers have successfully solved a problem in graph theory that has attracted attention from within the field. The team's research involves packing coloring, which deals with labelling parts of a graph to comply with certain rules and avoid specific conflicts.
A digital twin of the bladder has been developed to simulate normal and bladder outlet obstruction (BOO)-affected function. The model will help researchers better understand the connection between changes in BOO bladder wall structure and functionality, enabling them to develop new treatments and predict treatment success rates.
Researchers developed an AI foundation model for eye care that can identify sight-threatening eye diseases and predict general health. The model, RETFound, was trained on millions of eye scans from the NHS and outperforms existing state-of-the-art AI systems across complex clinical tasks.
A University of Houston study found that most of Houston's ozone exceedance is due to transported pollutants from the central and northern US, while local photochemistry contributes to elevated ozone production. The research highlights the importance of reducing emissions at the Houston Ship Channel to mitigate future ozone pollution.
A novel AI system developed by City University of Hong Kong improves predictive accuracy in dense traffic, reducing latency and increasing efficiency. QCNet achieves speed and accuracy in predicting road users' movements, even with long-term predictions, making autonomous driving safer and more human-like.
A team of scientists from Ames National Laboratory developed a new machine learning model that predicts Curie temperatures of new material combinations. This breakthrough discovery is crucial for designing high-performance magnets with reduced critical materials.
A study published in Radiology found that combining short- and long-term breast cancer risk models using artificial intelligence can improve cancer risk assessment. The combined model showed an overall improved risk assessment for both interval and long-term cancer detection, identifying women at high risk for breast cancer.
A hybrid system of electronic encoding and diffractive optical decoding transmits optical information with high fidelity through random, unknown diffusers. The system outperforms traditional approaches that only utilize a diffractive optical network or an electronic neural network for optical information transfer.
The WVU team evaluated Code Interpreter's features, finding it accessible to students but limited for scientists working with biological data. The plugin breaks down barriers for coding, but lacks internet access and parallel processing capabilities.
MIT researchers developed a new algorithm that can execute complex maneuvers like loops and rolls for tailsitter aircraft, enabling agile trajectories with fast-changing accelerations. The algorithm uses differential flatness to ensure feasibility and planning in real-time.
A new climate modeling method called ensemble boosting can simulate a large set of extreme but plausible heat waves, providing a worst-case scenario for planning and preparation. This method helps prepare for the potential loss of tens of thousands of lives in extreme heat waves.
The University of Texas at El Paso has been awarded a $5.3M grant from the Air Force Research Laboratory to enhance digital engineering training programs. The project will provide pre-professional experiences to 200 students and offer courses in digital engineering.
Aerodynamic researchers at University of Illinois create wind tunnel experiment to study internal boundary layers and their impact on flow behavior. They identify a new internal boundary layer that changes the flow's behavior, providing insights into aerodynamics physics and improving turbulence models for complex designs.
A study by Cornell University researchers found that current methods for measuring malaria parasite multiplication rates vastly overestimate the actual rates, which has significant implications for vaccine efficacy and understanding of drug resistance.
A team led by Dr. Zixiang Xiong at Texas A&M University aims to understand the fundamental limits of learned source coding, a machine learning-based data compression method. They hope to develop more powerful compression methods for efficient use of wireless communication and less energy consumption by mobile devices.
A new study by Paula Mayer proposes a practical tool for identifying areas of high human-bear conflict in the Abruzzo region. The model considers factors such as habitat suitability, migration corridors, and human-made food resources to inform local measures promoting coexistence.
Researchers used AI to accurately classify four subtypes of Parkinson's disease from patient-derived stem cells, with one subtype reaching an accuracy of 95%. The study suggests that personalized medicine and targeted drug discovery could be possible using this approach.
A new study found that a massive North Atlantic cooling event led to the disruption of early human occupation in Europe, with climate stress changing the course of early human history. The study used observational and modeling evidence to document the unprecedented climate anomaly.
Researchers aim to create machine learning tools that can analyze and quantify shape information from images, enabling more accurate diagnoses and improving patient care. A new family of deep neural networks, called DSNNs, will be developed to tackle AI's blind spot in image analysis.
Researchers have developed a new explainable AI model to reduce bias and enhance trust in machine learning-generated decisions. The Pattern Discovery and Disentanglement (PDD) model can predict medical results with rigorous statistics and explainable patterns, leading to more reliable diagnoses and better treatment recommendations.