Articles tagged with Artificial Intelligence
A team of computer scientists created 2,300 original sudoku puzzles and asked AI tools like OpenAI's ChatGPT to solve them. The results showed that while some AI models could solve easy sudokus, most struggled to provide accurate explanations, raising questions about the trustworthiness of AI-generated information.
A new blood test leverages AI to identify Lyme disease sooner and more accurately than current standard tests, with sensitivity and specificity over 90%. Additionally, an AI tool, Medicine-GPT, demonstrates promise in helping adolescents gather useful medical information, outperforming ChatGPT-4 on completeness and helpfulness.
New research highlights the transformative power of digital technology in shaping the e-patient era, enabling active participation and co-production in healthcare. Key technological innovations have built upon each other, providing patients with unprecedented tools for self-care and interaction with the healthcare system.
A machine learning model accurately predicted no-shows and late cancellations in primary care practices, with the gradient boost model achieving 85% accuracy for no-shows and 92% for late cancellations. Patient characteristics such as sex, age, and socioeconomic status played a significant role in missed appointments.
Mayo Clinic has deployed NVIDIA Blackwell infrastructure to accelerate pathomics, drug discovery, and precision medicine using generative AI. The new infrastructure enables faster pathology slide analysis and foundation model development, reducing work time from four weeks to one week.
A new study reveals that pedestrians are now walking faster and spending less time in public spaces. Researchers analyzed 40 years of video footage to find a 14% decline in people lingering in these areas.
A new AI model called RiboNN predicts translation efficiency of mRNA sequences, accelerating the development of mRNA therapeutics. The tool helps predict how much protein cells will produce, minimizing trial-and-error experimentation.
A special issue examines how AI can support human flourishing in various contexts, including healthcare, education, and digital interactions. The issue highlights the need for humane AI that balances technological advancement with human-centric values and psychological insight.
A new framework developed by Northwestern University and UCLA scientists integrates various water-related processes with a machine-learning model to predict landslide threats. The framework identifies three main pathways leading to landslides: intense rainfall, rain on already saturated soils, and melting snow or ice.
Researchers use AI to solve differential equations, such as Schrodinger's equation, for large-scale systems, improving efficiency and accuracy in fields like drug discovery and material design.
The team aims to deliver AI power directly to devices, improving resilience and speed in constrained environments. By processing data step-by-step across a network of devices, they can create a safe and adaptable system that can withstand attacks and extreme conditions.
A study published in PLOS One found that AI-generated music triggered greater physiological responses and was perceived as more emotionally stimulating compared to human-created music. The complexity of the prompt used also affected the impact of the music, with more complex prompts causing higher cognitive loads.
A new robotic slip-prevention method has been developed to improve robots' grip and handling of fragile or slippery objects. This bio-inspired approach allows robots to predict when an object might slip and adapt their movements in real-time, outperforming traditional strategies.
Answer ALS is launching a groundbreaking collaboration with Tulane University and the Pennington Biomedical Research Center to harness AI for ALS treatment discovery. The Louisiana AI Drug Development Infrastructure for ALS (LADDIA) will prioritize therapeutic targets using AI-driven insights from the Answer ALS' Neuromine Data Portal.
Researchers found that ChatGPT-4 performed better across demographic groups, while LLaVA showed significant sex-related biases in diagnosing skin diseases from medical images. The study emphasizes the need to address these biases to ensure AI models are safe and effective for all patients.
Two research teams will use AI to advance screening and treatment for cardiovascular disease, a leading cause of death in the US. The projects aim to improve health outcomes and save lives by detecting early signs of kidney and liver disease, as well as personalizing CPR techniques.
A global research team developed an AI model that can predict the death risk of hospitalized liver patients with cirrhosis, outperforming older methods. The model uses advanced machine learning tools to identify key risk factors, allowing healthcare providers to intensify care and make informed decisions.
Researchers at the University of Waterloo have created a tool called UnMarker that can remove any AI image watermark, highlighting the vulnerability of this technology to deepfakes. The team's discovery shows that even invisible encoded watermarks can be destroyed without knowing their design or internal parameters.
A novel AI-powered flood damage assessment using satellite imagery has been developed, achieving 74% of fully supervised performance with just 10% labeled data. The Simple Prior Attention Disaster Assessment Net (SPADANet) model improves recall by over 9% compared to existing models.
Researchers developed an AI-based screening tool using pangrams to detect Parkinson’s disease with nearly 86 percent accuracy. The web-based test analyzes voice recordings for subtle patterns linked to the neurodegenerative disease, identifying potential warning signs.
Research combines AI-generated disinformation with the repetitive truth effect to create persuasive content. A study found that while neither tactic convinced participants of conspiracy theories, AI-generated disinformation increased perceived consensus among both parties.
A University of Kansas study found that people rate corporate crises messages written by humans as more credible and trustworthy, regardless of the approach taken. However, the approach itself didn't vary between participants who read human or AI-written content.
The lab utilizes AI to automate and accelerate nuclear license applications, reducing review times by approximately 90% compared to traditional methods. This innovation aims to improve the efficiency of the licensing process while maintaining safety standards.
CORNETO enables researchers to extract molecular networks by combining experimental data with prior biological knowledge, helping to better understand cell health and disease mechanisms. The tool identifies shared pathways and patient-specific differences, a step toward personalised treatment strategies.
Researchers developed an AI tool to automatically generate HIV prevention and testing campaign messages for counties in the US, leveraging real-time social media data. The method was tested in three studies, showing that AI-selected messages were perceived as more actionable and effective by target audiences than control messages.
Researchers developed an AI model that accurately differentiates patients with various neurodegenerative disorders using 3D brain imaging data. The model achieved high accuracy, particularly for vascular dementia and Alzheimer's disease, by focusing on subcortical brain structures.
A new UNLV-led study uses AI to detect emerging virus variants in wastewater samples, outperforming existing methods. The algorithm can identify unique signatures for different virus variants with as few as two to five samples, significantly earlier than current methods.
Researchers developed CoSyn, a new approach to train open-source models using AI-generated scientific figures and charts. The resulting dataset, CoSyn-400K, includes over 400,000 synthetic images and 2.7 million sets of corresponding instructions. CoSyn-trained models match or outperform proprietary peers in various benchmark tests.
Researchers propose adaptations to overcome barriers to the implementation of autonomous AI agents in healthcare, including extending enforcement discretion policies and developing voluntary alternative pathways. Long-term solutions include regulating AI agents like medical professionals, with structured training processes for autonomy.
MIT engineers developed a versatile demonstration interface that allows users to teach robots new skills in three intuitive ways: remote control, physical manipulation, or demonstration. This innovation expands the type of users and 'teachers' who interact with robots, enabling robots to learn a wider set of skills.
Researchers found that deep neural networks exhibit absorbing phase transitions, a phenomenon observed in physical systems like forest fires. This discovery provides a unified framework describing how the signal propagates between layers of neurons, enabling prediction of trainability and generalizability.
The study explores the technological evolution of surgical robots and their current clinical applications. Surgical robots have addressed challenges in operating within small anatomical spaces, improving medical efficiency, but also increasing costs and complexity of operations.
The review proposes a systematic AI-driven framework for OLED material design, offering theoretical guidance and practical pathways for intelligent materials development. Machine learning models are used to predict core optoelectronic properties, enabling rapid and accurate property prediction while providing structural insights that g...
CodeSteer, a smart coach developed by MIT researchers, guides large language models (LLMs) to switch between text and code generation to solve complex problems. The system boosts the accuracy of LLMs on symbolic tasks like scheduling shipments in supply chains and multiplication problems.
The new compute roadmap aims to improve lives and livelihoods with advanced computing power. It includes investments in AI research, healthcare diagnostics, and renewable energy.
Researchers used an AI model to reassess a completed clinical trial for an Alzheimer's disease drug, finding it slowed cognitive decline by 46% in early-stage patients. The AI model predicts patient progression and allows for more precise selection of trial participants, potentially reducing the cost of developing new medicines.
A new poll suggests that older adults are divided on the use of artificial intelligence (AI), with some expressing curiosity and others skepticism. The study found that nearly all participants wanted clear labeling of AI-generated information, while also acknowledging its potential benefits for independent living. However, many express...
ADLM 2025 will tackle urgent topics like clinical AI integration and fake medical news, as well as leveraging genomics for personalized healthcare. The meeting also explores the health threat of plastics and microbiome medicine to combat diseases.
A study on public perception of physicians who use AI found that while effects are relatively small, they may be highly relevant due to the close link between trust in healthcare practitioners and subjective treatment outcomes. The integration of AI in healthcare has sparked certain reservations among the public.
Researchers created an 'encyclopedia of sugar utilization pathways' in 263 Bifidobacterium genomes. This resource helps predict which strains thrive in different conditions, enabling personalized probiotics to match children's lifestyles and dietary needs. The study suggests a potential solution to improving outcomes for preterm infants.
Researchers from the Weizmann Institute of Science and Intel Labs have developed new algorithms that allow AI developers to combine the power of different AI models, speeding up large language model performance by 1.5-2.8 times. This enables faster collaboration between models, reducing processing power costs and improving overall effi...
Georgia Tech is building a new artificial intelligence-centered supercomputer called Nexus, which will integrate cutting-edge hardware and software services to unify scientific research workflows. NCSA is lending its expertise to assist in the development of this national-scale computational resource.
The Idaho National Laboratory is using Microsoft's Azure cloud and artificial intelligence to automate the nuclear permitting and licensing process. The tool generates reports required for construction permits and operating licenses, streamlining efficiency and accelerating deployment of advanced nuclear technologies.
Scientists use human-AI collaboration to tackle complex questions in condensed matter physics, leveraging machine learning algorithms to identify patterns in simulation data. This approach successfully models the behavior of frustrated magnets and sheds light on quantum computing and gravity.
Researchers have developed a groundbreaking technique that maps temperature in three dimensions within biological tissue using invisible light and artificial intelligence. This new technology has the potential to improve early disease detection and treatment monitoring without the need for costly or invasive imaging technologies.
A new approach by MIT researchers allows scientists to efficiently estimate how combinations of treatments will affect a group, enabling fewer costly experiments while gathering more accurate data. The framework considers the scenario where all treatments are assigned in parallel and controls the outcome by adjusting treatment rates.
A NUS-led global study reveals that 30-minute functional MRI scans deliver up to 22% in cost savings while retaining or improving prediction accuracy. This finding could reshape how researchers design neuroscience and mental health studies, especially for hard-to-recruit populations.
SmartCatch, an AI tool developed by WorldFish, has won the AI for Planet Prize at the 2025 AI for Good Global Summit. The tool helps small-scale fishers record their catches with a simple mobile phone photo, aiming to make catch reporting faster and more accurate in coastal areas with limited internet connectivity.
Stanford researchers introduce a cost-effective way to evaluate AI language models, accounting for question difficulty to ensure fair comparisons. The new approach reduces costs by half and enables rapid progress in the field of artificial intelligence.
Researchers from Memorial Sloan Kettering Cancer Center developed a new approach to induce mutations in resistant cancer cells, making them sensitive to immunotherapy. The study showed positive outcomes in patients over 90 with lung cancer who underwent surgery.
Researchers created a digital twin to detect disease risk, initiate preventive treatment, and simulate effective treatments. The model uses extensive human data from the Human Phenotype Project to identify deviations from expected physiological changes in 17 body systems.
An AI model accurately detected tumor locations and outperformed benchmark models in detecting breast cancer, according to a study published in Radiology. The model learned a robust representation of benign cases to better identify abnormal malignancies, even with underrepresented data.
Researchers aim to create automated tools to safely translate legacy C codebases into memory-safe Rust without sacrificing performance or functionality. The project, backed by DARPA, uses formal methods and AI techniques to develop accurate translations of complex C code.
The Hong Kong Polytechnic University has secured funding from the Research Grants Council to develop a cost-effective and sustainable collaborative Generative AI (Co-GenAI) model. The project aims to democratize AI development by reducing centralized computational demand and enabling broader participation in GenAI research and deployment.
A new deep learning model enhances handheld 3D medical imaging by automatically tracking transducer motion without external sensors. The model produces more realistic 3D US images and can reconstruct blood vessel structures using ultrasound and photoacoustic data.
Researchers found that specific gaze sequences, including looking at an object and making eye contact, are crucial for effective communication. This discovery has implications for social robot development and non-verbal communication training.
Researchers at the University of Pennsylvania used AI-powered screen of global venom libraries to discover dozens of promising drug candidates. The top peptide candidates are being improved through medicinal-chemistry tweaks.
Researchers developed Localizatome to study oxidative stress-related changes in protein localization. The database provides comprehensive information on subcellular protein localization and dynamic localization changes under stress.
The EVOaware project aims to develop an innovative platform that addresses tumour resistance to therapies by using advanced tissue imaging technologies and integrating genetic screening, lineage tracing, and spatial omics techniques. This platform has the potential to accelerate the discovery and development of new cancer therapies.
A recent survey found that nearly two-thirds of U.S. adults consider AI-generated summaries to be somewhat or very reliable, while another 45% say they sometimes give them the answer they need. However, half of respondents are not comfortable with healthcare providers using AI tools.