Research applying artificial intelligence across science and engineering. Covers machine learning, sensing and quality control, immune and health prediction, and 5G networks.
The authors propose a spectrum of clinical autonomy, ranging from advisory tools to navigator systems that work with minimal human oversight, to enable early prediction and prevention of disease. However, challenges related to clinical validation, integration, data quality, and regulatory approval limit widespread deployment.
A new framework enables AI agents to collaborate at scale, integrating reasoning into communication. The framework has three capabilities: intent interpretation, automated formulation and optimization, and proactive foresight, paving the way for 7G wireless networks with collective intelligence.
The new center aims to better understand diseases and treatment response by combining human genetics and microbiome research with AI. Researchers will investigate how genetic variation and the gut microbiome jointly shape disease progression and treatment response.
The 13th Heidelberg Laureate Forum brings together 200 young researchers in mathematics and computer science with 21 distinguished laureates, including 2026 Fields Medal and Abel Prize winners. The event features a range of lectures, panel discussions, and a 'Next Gen Session' for young researchers to showcase their work.
Scientists at Scripps Research developed an AI model called ECG-CLIP that improves heart disease detection and prediction with less labeled data. The model outperforms existing tools in various clinical tasks, including disease detection, prediction, and adverse health outcomes, with better performance in settings with limited data.
Researchers aim to develop new magnets that reduce American reliance on supply-vulnerable foreign sources by finding alternatives to critical minerals. The UH-led team will use AI to design and manufacture next-generation permanent magnets, with the goal of surpassing industry-standard materials like neodymium iron boron.
Rice University bioengineer Michael King joins the Texas Cancer Institute's AI advisory panel to explore AI's potential in cancer research and prevention. The panel aims to ensure responsible and ethical use of AI in biomedical research, with King's expertise in cancer bioengineering playing a key role.
A new AI framework, AdvWT, uses realistic fading, cracks, and corrosion to test vulnerabilities in AI vision systems. It achieves near-perfect attack success rates on lightweight CNNs and transformer-based models, and can also be used to restore naturally damaged traffic signs.
A cross-sectional study suggests that seeking emotional support from generative AI is a unique marker of psychological distress in children and adolescents. Clinical frameworks and digital literacy programs should distinguish between functional AI assistance and algorithmic interfaces as a digital refuge for emotional needs.
Researchers found that AI-generated ads can be ineffective when the technology is not matched to the advertisement's theme and intent. Marketers should use AI to enhance campaign ideas, not as a shortcut. The study suggests a stronger match between AI and ad themes can increase consumer engagement.
A new review suggests that artificial intelligence could help scientists develop more precise and mechanism-guided biochar management strategies for acidic soils. Researchers should distinguish between organic and inorganic sources of alkalinity in biochar to improve predictive models.
NEW HORIZON PRESS LIMITED participated in ACS Fall 2026, connecting with researchers and scholars to introduce its academic journals and publishing initiatives. The meeting highlighted advances in chemistry, materials science, and environmental science, with a focus on sustainability and interdisciplinary research.
A KAIST team uses AI to identify optimal material recipe for 3D-printable, highly stretchable material. The material printed reliably on a DLP 3D printer and showed high stretchability, extending to over six times its original length.
Researchers have developed a novel AlphaFold-based method that introduces a repulsive force between predicted structures, allowing for the sampling of multiple conformational states. This enables the prediction of diverse protein conformations rapidly and accurately, with potential applications in drug design and protein engineering.
Researchers at KAIST have developed a low-cost smartphone-based technology to detect hidden cameras by analyzing reflections from objects. The technology, called SweepLED, uses deep learning-based analysis to distinguish camera lenses from ordinary objects, achieving 94% detection accuracy. The technology has potential to be developed ...
A new generation of intelligent decision support methods and software tools will be developed to enhance supply chain resilience, optimize healthcare operations, and strengthen emergency response coordination. The HOPE project aims to create a lasting area of excellence in human-centered AI, serving as a trusted partner for government,...
Digital tools are transforming clinical research, public health readiness, healthcare institutions, and consumer choices. Citizen science initiatives using mobile apps are also growing, with three types of citizen science identified: with-the-people, by-the-people, and for-the-people.
Researchers developed fluorescent molecules that allow imaging of DNA inside living cells at unprecedented resolution, and in preserved cells, close to the width of the double helix. The team tested the probes on cancer patients and found DNA to be noticeably looser and more spread out in tumours.
Researchers have created a high-resolution functional map of human immune cells, revealing intricate circuits that govern health and disease. The dataset provides a powerful framework for designing cancer immunotherapies and treating autoimmune conditions, and serves as a foundation for AI models of biology.
Researchers develop adaptive multi-expert framework for dynamic 3D reconstruction, combining strengths of multiple motion representations to improve reconstruction quality. The framework leverages complementary experts to handle heterogeneous dynamics, enabling more accurate reconstruction of complex scenes.
Researchers develop a novel framework, LL-Refiner, to enhance high-resolution images in poor lighting conditions, outperforming state-of-the-art techniques. The framework uses a coarse enhancement stage to guide the recovery of fine details, resulting in improved visual quality and performance in downstream computer-vision tasks.
A new study suggests that machine learning models using first-trimester pregnancy data can identify women and babies at risk of serious health problems earlier and more accurately than existing early risk assessment approaches. The models generally outperformed the current methods in Sweden, Chile, and Singapore, highlighting the poten...
Researchers trained AI on stick insect walking cycle to find optimal walking strategy, resulting in a six-legged robot that can navigate treacherous terrain and adapt to missing limbs. The approach allows for cheaper and faster robot production, enabling potential disaster response applications.
Researchers found that sector labels in the S&P 500 only partially capture a company's financial situation. The study used machine learning models to group firms by financial similarity, revealing nine economically interpretable clusters that generally showed lower internal dispersion than conventional sectors.
A new AI-based diagnosis tool can detect hypertension and diabetes from a single facial video, with high accuracy and potentially reaching far more people than traditional screening methods. The algorithm can estimate blood pressure from a facial video alone, without a cuff.
A WVU researcher is working to make AI systems more transparent about their uncertainty, to prevent misinformation and improve trust in high-stakes fields like healthcare. The goal is for AI systems to identify when they're unsure and ask questions or provide more nuanced responses.
A USF researcher explores whether human oversight is meaningful when AI has already shaped the information reaching a decision-maker, raising questions about AI ethics and responsibility. The researcher's findings highlight the need for greater awareness of the systems in which AI research and use are embedded.
Four assistant professors, Yahong Yang, Sammy Luo, Lebing Chen, and Kunyan Zhang, join Binghamton University as Simons Empire Faculty Fellows, bringing expertise in quantum materials and artificial intelligence. Their research focuses on developing new technologies, including energy-efficient systems and next-generation sensing platforms.
A new tool called Socratic Challenger uses AI to help undergraduate students cultivate critical thinking skills in research question development. The framework, consisting of eight steps, including AI-enabled tasks, enables students to think critically about research and develop meaningful research questions.
MIT researchers developed a framework, CrysVCD, to generate stable materials with desired properties, reducing the need for extensive screening. The approach improves material stability by 70% and supports the creation of high-performance materials, such as computer chips and data centers.
Scripps Research will establish an open-access autonomous chemistry laboratory with a $19.5 million NSF award, advancing AI-driven discovery and education in chemistry. The lab will utilize AI and automation to streamline chemical reaction discovery and optimization, making it more efficient and accessible to a broader community.
A team of Korean researchers has created a miniaturized wireless brain implant that can deliver drugs and light to precisely modulate targeted neurons remotely. The device overcomes distance and location constraints, enabling long-term studies of brain disorders and therapeutic devices.
Researchers found that 21 leading open-weight AI models can be modified to bypass safety protections, raising concerns about mass disinformation campaigns and hazardous chemical production. The study's lead author notes that the weaknesses may not be unique to open models, highlighting the need for stronger security systems.
A new study from the University of British Columbia finds that AI cannot replace traditional language learning, but a hybrid approach combining corpora and AI can be effective. The study suggests that educators use AI to verify student answers and improve accuracy, while still relying on traditional corpora for teaching collocations.
Researchers propose a new AI framework, Generative Electrochemical Intelligence, to accelerate the discovery and development of electrochemical energy technologies. The framework combines generative AI with automated robotic experimentation to create a closed-loop system that can generate new ideas, test them, and learn from feedback.
DigBat brings together solid-state electrolyte data, simulations, machine learning, and AI to support battery materials research, providing a clearer view of the solid-state electrolyte landscape. Researchers can compare experimental and computational data, build machine-learning models, and gain insight from the data.
Researchers create unique 'artificial fingerprints' using nanoparticles that can be authenticated with smartphone flashlight and laser pointer. The technology has potential applications in anti-counterfeiting and electronic device authentication.
A new brain-inspired algorithm, Spi-Fly, demonstrates promise for achieving practical applications in scent classification, particularly in scenarios with limited training data. The algorithm shows accurate classification of scents and can learn with few-shot and continual learning methods, making it suitable for real-world applications.
A new blood-based approach developed by Kumamoto University researchers detects breast cancer recurrence by analyzing nucleosome structure in circulating DNA. The study identified genomic regions associated with treatment resistance and recurrence, promising a low-invasive monitoring method for patients.
The ST-NUS HELIX Corporate Lab aims to develop new generative and embodied AI use cases at the edge through system-to-silicon innovation, reducing energy consumption and improving performance. Researchers will focus on memory-centric architecture, innovative in-memory computing, and scalable compute-and-memory systems.
Academic fraud, aging, and healthcare automation are addressed through digital health innovations, including forensic scientometrics and intelligent monitoring. These solutions aim to improve working conditions, patient care, and healthcare access, addressing systemic challenges and transforming clinical workflows.
A cross-sectional study found that adolescents increasingly use AI chatbots to discuss mental health concerns, with significant implications for young people's well-being. The study's findings highlight the need for increased awareness and support for adolescents' mental health needs.
A new smartphone app called Mobilio uses AI, machine learning, and personalized audio cues to provide turn-by-turn directions, path guidance, and obstacle avoidance for people with blindness or low vision. The app completed outdoor navigation tasks 13% faster and reduced obstacle contact by 41% compared to Google Maps and a white cane.
MIT engineers develop a tool that generates plausible extreme events and worst-case scenarios without relying on extreme data, enabling planners to prepare for unprecedented scenarios. The algorithm takes a statistical approach to learn from available data, excluding implausible weather scenarios, and projects how extreme events might ...
Researchers integrate AI into local monitoring sensors to track ecosystem health in near real-time, reducing delays of months to years. The project enables faster release of accessible flux data, helping scientists understand ecosystem responses to change and inform land management decisions.
Researchers used AI to predict genes leading to hematopoietic stem cell aging, finding a regulatory hub in the form of gene-regulating factor Pbx1. Aged stem cells activate two gene programs, one preserving immaturity and the other preparing cells for platelet production.
Scientists in the University of California San Diego laboratory used AI to decipher the 'initiator' DNA sequence, which is responsible for gene activation. The researchers found that about 60% of human genes contain the initiator, enabling the prediction of DNA mutations that can lead to various disorders.
Two OU-led research teams have been awarded nearly $1.4 million in funding under the DOE's Genesis Mission to develop AI for enhanced geothermal systems and quantum computing. OU's projects align with the university's impact-centers research priorities and address the Genesis Mission's pillars of Energy Dominance and Discovery Science.
Researchers developed a technique to assess the reliability of medical imaging tools, which can be used to evaluate quantitative imaging methods and build confidence in these technologies. The technique, called NGSE-Corr, was shown to accurately rank imaging methods for 91% of trials and identify the most precise method for 95% of trials.
This theme issue explores clinical efficacy, seamless support systems, and ethical AI integration for home-based care. Research topics include 'invisible' monitoring, computer vision, socio-technical drivers of AI adoption, and clinical implementation.
The PolyU research team has developed TRUECAM, an integrated AI framework that ensures data and model trustworthiness in cancer diagnosis. The framework can assess the level of an AI's confidence in its diagnostic outputs and proactively prompts pathologists to review cases when uncertainty is high.
McGill researchers have developed a more energy-efficient method for building AI systems that can measure and indicate their own uncertainty. This approach cuts memory and training costs while maintaining strong predictive performance. The researchers aim to make reliable, uncertainty-aware AI practical for large and complex systems.
A new navigation system for cyborg insects combines AI-based real-time terrain recognition with the cockroach's natural climbing ability, enabling faster and more efficient navigation. This approach allows the insects to traverse obstacles, climb walls, and cross holes with reduced detours and steering stimulation.
The University of Hong Kong has pioneered 'RoboDojo', a unified benchmarking platform evaluating robotic manipulation across simulated and physical environments. The platform's initial findings reveal a significant performance gap between current robotic systems and human capabilities.
A novel AI model has been developed that can recognize yoga poses with high accuracy, paving the way for more effective digital coaching tools and movement-monitoring applications. The model achieved accuracy levels of over 93% during testing, significantly outperforming previous models.
Mount Sinai scientists reveal how the brain represents leader and follower roles through prefrontal cortex activity, and create an AI that decodes hidden goals behind teamwork. The study shows that leadership is an asymmetric yet bidirectional partnership, and that followers play a crucial role in maintaining cooperation.
A study published in PNAS found that the number of AI agents in a group affects their collective decisions, sometimes amplifying existing biases or even inventing new ones. As the group size increases, the agents' preferences become more predictable, but the point at which this happens varies depending on the model and task.
A new analysis found that only 3 out of 1,357 AI medical devices authorized by the US FDA were tested on patient outcomes, while most were only evaluated on substantial equivalence to existing devices. The study suggests that existing policies for AI device authorization should be redesigned to prioritize clinical effectiveness.
Research from the University of Birmingham and other institutions found that interacting with AI-powered customer service robots can reinforce or alter a consumer's self-perception. The study explores how mirroring and mimicry can lead to a 'robotoid humanness' where consumers become more like robots, raising ethical concerns.
A machine learning pipeline developed by LMU researchers reveals notable imbalances in global health aid allocation, with non-communicable diseases receiving only 2.5% of disease-specific aid funding despite making up 60% of the global disease burden. The study's findings suggest that health aid is being directed to the wrong areas, an...