Articles tagged with Artificial Intelligence
Researchers developed an AI model that can detect several neurodegenerative diseases, including Alzheimer's and Parkinson's, from a single blood sample. The model uses protein measurements from over 17,000 patients and outperforms previous models.
New research warns that generative AI can reduce feedback to a transactional exercise if not guided by principles of care and connection. Instead, a 'care-full' approach to feedback is needed to promote meaningful learning.
The conference will feature keynote dialogues, fireside chats, and panel discussions on key themes such as AI and education, law and regulation, public governance, and global cooperation. Distinguished participants include senior representatives from international organisations and public policy institutions.
HeapGrasp uses RGB images to analyze object silhouettes and estimate its 3D shape, reducing the need for depth information. The approach achieves high accuracy while minimizing camera movement and execution time.
Researchers at Tohoku University developed a method to transform regular surfaces into keyboards for AR/MR technology, using the blanching phenomenon to detect fingertip contact. The system requires no special sensors or devices and allows users to rest their fingers on the surface during operation, promoting comfortable interaction.
A new AI-based system analyzing blood vessels at the back of the eye has been shown to identify people at risk of heart problems with strong correlation. The system, called CLAiR, received Breakthrough Device designation and demonstrates promise as a noninvasive screening method.
A Mayo Clinic study found that AI-derived heart fat measurement can predict cardiovascular events more accurately when combined with traditional risk assessment approaches. The measurement significantly improved long-term risk prediction, especially among patients at low-risk categories.
A recent study published in Gait & Posture found that analyzing a person's walk and getting up from a chair can identify elevated depression and anxiety symptoms. The researchers used machine-learning models trained on data from participants' movements combined with information about their mental state, achieving high accuracy rates.
Simon Fraser University (SFU) and Queen's University are partnering to design and build a national sovereign, secure and sustainable high-performance supercomputing system. The collaboration aims to strengthen Canada's sovereign AI supercomputing capacity, support industry innovation, and ensure data security.
A new study presents a framework combining biochar engineering with artificial intelligence to design next-generation materials tailored for specific pollutants. The work highlights how advanced data-driven approaches can accelerate the development of sustainable water treatment technologies.
A newly developed AI tool called EvORanker analyzes genetic patterns across over 1,000 species to identify the cause of rare diseases. In clinical testing, it successfully identified the disease-causing gene in nearly 70% of cases, offering new hope for treatment and closure.
Researchers developed a machine learning model to predict 28-day mortality in patients with sepsis complicated by acute respiratory failure. The model demonstrated strong discrimination for predicting mortality, with key variables including oxygenation indices, serum albumin levels, and disease severity scores.
New research from the University of East London suggests that machine learning can improve the accuracy and nuance of personality tests like DISC assessment. Using over 1,000 participants, researchers achieved accuracy rates of over 93% in predicting personality types and identified four clear clusters with subtle overlaps.
A new study proposes a learning architecture that integrates educational philosophy with AI-driven design, aiming to transform assessment into an ongoing process of reflection. The system prioritizes human judgment and interpretation over standardized metrics, enabling educators to build adaptive and interpretable feedback systems.
Researchers developed AI models to analyze conversations between children with cancer and their caregivers, detecting severe symptoms that require extra support. More complex prompting strategies outperformed simpler ones in accurately identifying symptom severity and its impact.
The study found that AI-enhanced single-shot cine MRI produces better image quality compared to conventional cine MRI, particularly in participants with arrhythmia. The technique demonstrated a 100% success rate for image acquisition, outperforming conventional cine sequences.
Researchers developed an AI tool called PathogenFinder2 that can detect harmful bacteria before they infect humans. The tool uses protein language models and has been shown to significantly improve the detection of bacterial threats.
A new model combines multiple ways of analysing 3D data, integrating local and global perspectives to interpret complex environments more reliably. The system improves detection of small or partially visible objects in real-world situations, enhancing safety in autonomous systems.
A study by Stanford University researchers found that AI large language models are prone to sycophancy when providing advice on interpersonal dilemmas. The models often affirmed users' choices, even in harmful or illegal scenarios. Participants who interacted with sycophantic AIs reported increased self-centeredness and moral dogmatism.
Researchers at MIT have developed an ultrasound wristband that precisely tracks hand movements, allowing users to control a robotic hand or manipulate virtual objects. The device produces high-quality images of the wrist's muscles and tendons, which are then translated into specific hand positions, enabling precise movement control.
Digital Science acquires Ontopic, a pioneer in Virtual Knowledge Graph technology, to democratize research data and provide enterprise-grade AI solutions. The acquisition enables users to build and access Knowledge Graphs directly from existing data sources.
Gladstone Institutes has secured over 105,000 square feet of future laboratory space in a newly constructed building, empowering its scientists to create medicines of the future. The new space will be home to approximately 300 scientists across 20 labs, equipped with state-of-the-art equipment and computational abilities.
ChatGPT versions showed high rates of inappropriate responses, particularly among economically disadvantaged individuals with psychosis risk. GPT-5 performed better than the free product version in the across-version analysis.
Researchers developed an AI model to create highly photorealistic 3D reconstructions of ground-level damage after earthquakes. The LoRA-Enhanced Ground-view Generation diffusion model can recognize complex visual patterns and predict where structures may be damaged, even in densely populated urban areas.
The UJI's GRAPE group proposes a paradigm shift in language teaching with a computer tool that analyzes verbal and visual elements. The software, GRAPE-MARS, enables fast labeling of video files and incorporates AI tools for efficient analysis.
The International Telecommunication Union (ITU) will host the seventh AI for Good Global Summit from 7 to 10 July 2026 at Geneva’s Palexpo convention centre. The summit aims to guide the future of artificial intelligence and unlock its potential to serve humanity.
The UOC study examines how technological initiatives promote governance models based on private digital identity and biometric data systems. It focuses on Worldcoin, a project that verifies users' humanity through iris scanning in exchange for a digital identity certificate.
Researchers from SDSU discovered surprising similarities among ancient writing systems from Africa and the Caucasus region. The study suggests the Armenian alphabet may be more closely related to the ancient Ethiopic writing system than previously thought, revealing possible cultural contact and influence between regions.
Researchers at Worcester Polytechnic Institute developed a palm-sized aerial robot that uses ultrasound sensors and AI to navigate through fog, smoke, and other difficult conditions. The drone achieved a success rate of 72% to 100% in navigating challenging courses during 180 tests.
Researchers discovered that stroke survivors with severe motor impairment show signs of brain 'youthfulness' in undamaged regions, suggesting the brain's ability to adapt and compensate. This finding provides new insights into neuroplasticity and may guide personalized rehabilitation strategies.
A new generative AI framework predicts future urban layouts considering multiple factors. The Memory-aware Multi-Conditional generation Network (MMCN) offers a novel approach to modeling complex urban evolution, providing a powerful tool for sustainable urban development.
Researchers at Dartmouth College developed a mathematical framework to map students' conceptual knowledge from short multiple-choice quizzes, revealing peaks of mastery and valleys of struggle. The technique could enable personalized learning, AI tutoring systems, and more efficient feedback.
A recent study published in PNAS Nexus found that large language models (LLMs) produce homogeneous creative outputs, regardless of the specific model used. This suggests that LLMs as a whole, rather than individual models, may be contributing to the lack of diversity in their responses.
A recent study proposes educational reconfiguration as the key to rebuilding trust and legitimacy in the age of artificial intelligence. The research identifies three singular transformations driven by AI and calls for a deeper structural transformation, focusing on cultivating civic literacy and human-AI collaboration.
A study found that deepfake X-rays can be indistinguishable from authentic images, posing a risk to patient diagnoses and digital medical records. Radiologists and AI models struggled to detect synthetic images, with accuracy ranging from 58% to 92%. Experts recommend implementing advanced digital safeguards to prevent tampering.
The American AI Jobs Risk Index projects 9.3 million US jobs at risk of displacement in the next 2-5 years, with a midpoint of approximately $757 billion in annual income loss. High-earning knowledge workers and occupations in Information, Finance, and Professional Services are most vulnerable to AI-driven job displacement.
Researchers discovered that key weight parameters contribute to both performance and data-privacy vulnerabilities in AI neural networks. They developed a technique to balance these aspects, achieving better results than existing methods in defending against membership inference attacks.
A team of MIT researchers has created a framework to design AI systems that are more collaborative and forthcoming about uncertainty in medical diagnoses. The system, which includes an Epistemic Virtue Score module, encourages users to gather additional information when diagnoses are uncertain.
A new AI pipeline called RAVEN has validated over 100 exoplanets, including 31 newly detected planets, using NASA's TESS data. The study found that around 9-10% of Sun-like stars host a close-in planet, and 'Neptunian desert' planets occur around just 0.08% of Sun-like stars.
Researchers use machine learning to characterize recovery processes, predict outcomes, and optimize strategies for restoring critical infrastructure systems. The review highlights the potential of reinforcement learning to identify effective repair sequences.
The article analyzes the legal and ethical complexities of explaining AI decisions to patients, highlighting significant hurdles such as interpretability trade-off, automation bias, and literacy barrier. It calls for co-design partnerships, institutional support, and standards for comprehension to bridge this gap.
The integration of AI in pediatric surgical care is transforming healthcare, but its application to children poses unique ethical challenges. Key findings highlight the need for transparent and accountable AI systems that prioritize patient safety, autonomy, and fairness.
Researchers at North Carolina State University have identified key components in large language models that ensure safe responses to user queries. They've developed a new technique to improve LLM safety while minimizing the alignment tax, which allows AI systems to provide safe responses without affecting performance.
A Cornell research team studied the integration of MLB's Automated Ball-Strike System (ABS), which uses AI to track pitches in real time. The researchers found that implementing technology into baseball requires complex consensus-building and sense-making among stakeholders, ultimately leading to a more accurate system.
Researchers created a new method combining scientific tests and artificial intelligence to differentiate recycled plastic from new plastic. The tool, developed by University at Buffalo researchers, can analyze samples and predict the percentage of recycled content with over 97% accuracy.
A landmark report by researchers at the University of Reading warns that weak legal rules on AI avatars are preventing UK businesses from accessing their full growth potential. The report proposes four reforms to ensure responsible use of AI avatar technology, including giving people a legal right over their own face and voice.
A new AI-powered method predicts peak oxygen consumption with high accuracy, identifying high-risk patients with advanced heart failure. The study's multi-modal model uses cardiac ultrasound images and electronic health records to detect this critical measure.
Researchers developed a new type of nanoelectronic device mimicking the human brain's efficient neuron connections, reducing energy consumption for AI systems. The hafnium-based devices achieve switching currents millions of times lower than conventional devices and store programmed states for around a day.
Researchers at Tufts University developed a proof-of-concept for efficient AI systems using neuro-symbolic AI, reducing power consumption by 100 times while providing accurate results. The approach uses symbolic reasoning to limit trial and error during learning, leading to faster completion and significant energy savings.
A clinical decision support system using AI-assisted imaging analysis and evidence-based treatment recommendations showed better quality care and long-term outcomes for patients with acute ischemic stroke. The tool reduced new vascular events by 26% at three months and maintained the reduction at 12 months.
Researchers used AI to analyze skin lesion changes in mastocytosis patients treated with avapritinib, finding a 36.6% reduction in lesion area and normalized skin color. AI can aid dermatologists in assessing small lesions, but human expertise remains essential.
Researchers at MIT have developed a new technique using generative artificial intelligence models to overcome the precision bottleneck in wireless vision systems. The method produces more accurate shape reconstructions, which could improve robots' ability to grasp and manipulate objects blocked from view.
A new benchmarking study shows that AI coding tools struggle with structured outputs in software development, with even advanced models achieving only 75% accuracy. This highlights the need for human oversight and suggests that AI systems are not yet reliable enough to operate without human supervision.
A team of computer scientists developed an algorithm that mimics bird flocking to help AI produce more reliable summaries of long documents. The framework reduces repetition and preserves key points, resulting in more accurate and concise summaries.
Scientists have added millions of protein complex structures to the AlphaFold Database, shedding light on how proteins interact. The dataset prioritizes human health and disease research, enabling researchers to test, refine, and build upon it.
A new AI model has been developed to predict floods with greater accuracy than traditional methods. The knowledge-guided approach combines elements of physics-based models with machine-learning techniques to automatically learn the state of a river's watershed, eliminating the need for manual recalibration.
A specialized AI model, AMIR-GPT, has been developed to improve radiology guideline alignment, outperforming general purpose models in 33.3% of test responses. However, the model's performance varied across performance bands, and qualitative review revealed limitations, such as omissions and deviations from standard recommendations.
Researchers developed a predictive model called Programmer Flow by Information Scent for teams (PFIS-T) that incorporates both teammates' activity and communication to predict a user's next action. The model achieved an accuracy of 81.5% in predicting team navigations, improving accuracy by up to 16.7% over previous models.
The new framework groups stations with similar hydrological behavior, reducing computational cost while maintaining high predictive accuracy. This approach enables scalable, data-efficient AI systems for water level forecasting, supporting flood early-warning systems, optimized reservoir and irrigation management, and improved decision...
ToxIndex integrates AI agents to access and orchestrate toxicological resources, providing comprehensive risk assessments in hours. The platform addresses a critical need in chemical and drug safety, meeting the vision of the 2007 NRC report for 21st-century safety testing.