Research applying artificial intelligence across science and engineering. Covers machine learning, sensing and quality control, immune and health prediction, and 5G networks.
Research in SLAS Technology Vol. 40 explores smartphone glucose sensing and RNA-based therapeutics in Crohn's disease. The publication showcases innovative technologies and scientific advancements in life sciences discovery and development.
Researchers at Chalmers University of Technology created an AI scientist capable of generating scientific hypotheses, designing experiments and interpreting results. The AI system, Eve, identified promising biological questions, recommended experiments and evaluated outcomes, and iteratively refined its understanding based on new evide...
A new AI tool, Tessera, has been developed to map smallholder crops in Senegal with high accuracy, providing essential data for food security planning. The technology, trained on satellite images, outperformed existing methods and can be used to guide support decisions, helping vulnerable populations.
Large language models (LLMs) evaluate humans based on cooperation and social norms, with most models favoring cooperation and punishing bad behavior. The LLMs' judgements often depend on the recipient's gender and cultural background, and may be influenced by prompting interventions.
A closed-loop modeling and assessment framework for aerospace engineering managers' AI competency is proposed, capturing governance priorities and multi-attribute interdependencies. The framework, comprising five dimensions and twenty attributes, provides a traceable decision-support tool for selection, performance management, and care...
A real-world study of 8,391 patients found that autonomous AI could free up enough clinical capacity to provide over 8,500 additional face-to-face dermatology appointments across two UK hospitals over 16 months. The technology identifies patients with benign lesions, enabling safe management without specialist review.
Researchers developed a label-free platform combining microfluidic cell sorting with AI to enrich and identify rare CTCs. The integrated strategy uses inertial microfluidic enrichment and YOLOv8-based deep learning for bright-field image recognition, achieving 96.0% accuracy in distinguishing tumor cells.
A new framework, PhysMat AI, integrates physical knowledge into AI for materials discovery, enabling more interpretable and testable predictions. This approach helps move materials discovery beyond correlation-based prediction toward reasoning based on physical principles.
A new AI tool, ChromAgeNet, analyzes 3D chromatin organization in blood stem cells to identify age-related changes, which can inform rejuvenation strategies. The model outperforms previous methods, revealing subtle changes in nuclear architecture that can be used to detect age-associated states.
Researchers found that AI chatbots give users a uniform range of information, similar to a conventional web search, but with a narrower scope. The study's authors warn of 'knowledge collapse' as language models become increasingly trained on AI-generated text, potentially reducing diversity and nuance.
Researchers at MIT have found a way to stabilize lipid nanoparticles used to deliver RNA vaccines, making them more heat-resistant. This breakthrough could allow for wider distribution and enable novel administration methods like microneedle patches.
Researchers developed a self-powered, flexible neuromorphic sensing platform that mimics human tactile perception, demonstrating hierarchical memory processes and spike-rate-dependent plasticity. The device operates entirely without an external power source, converting mechanical stimuli into electrical signals.
Researchers have developed a technology that stacks semiconductor devices with different response speeds to process fast and slow signal changes together. The technology could be applied to small devices that analyze movement and physiological signals, enabling the recognition of changes in motion over time.
Researchers developed a method called Concentration-Dependent Yield Analysis (CYAN) to connect reaction optimization and kinetic analysis. CYAN extracts kinetic information from yield data, estimating reaction speeds without requiring separate experiments. This approach helps chemists design complex high-yield reactions.
A new study evaluated the ethical properties of large language models (LLMs) in generating advice for urban design and health, finding that they avoid harm but lack consistency on community participation and human oversight. The study suggests that LLMs could serve as an initial input for urban designers, but should not replace profess...
A new study by Ateneo researchers explores the cost-effectiveness of AI-assisted chest radiograph interpretation in isolated and disadvantaged Filipino communities. The study finds that AI-assisted interpretation is estimated to be more cost-effective than manual interpretation, with a potential annual cost savings of Php 877,330. The ...
Researchers at NUS CDE have developed a reconfigurable transistor that can switch between filtering image data and performing AI network functions, reducing energy consumption and improving accuracy in handwritten-digit recognition simulations. The device uses a spiking neural network-in-logic architecture to selectively pass relevant ...
A recent study from Georgetown University and the University of Washington suggests AI-generated summaries can manipulate people's memory, even when accuracy is assured. Participants who read misleading AI summaries were significantly less likely to accurately recall the original event, according to the study.
Researchers developed an AI approach to predict glioblastoma recurrence, allowing for targeted treatments before the cancer becomes visible on MRI. The tool uses microscopic images of fresh, unprocessed tissue and scored based on tumor infiltration, with an accuracy of predicting recurrence within 5-10 millimeters of the sampled tissue.
A new AI tool predicts organ failure in acute pancreatitis with high accuracy, using multiphase CT imaging and outperforming standard scoring systems. The model achieves early automated prediction, with 55% of cases predicted at least 3 hours in advance.
The AI model achieves higher specificity while preserving sensitivity in breast cancer risk classification, reducing false-positive classifications and supporting its potential use as an adjunctive tool for refining positive imaging findings.
Researchers developed an innovative approach to identify insect sex pheromones by analyzing olfactory receptors, which can help control pest populations and improve crop protection.
Mizzou researchers have developed an AI tool called MeLSI that can identify specific microbes driving key biological changes in the gut microbiome, which may indicate early warning signs of disease. The tool has shown promising results in detecting meaningful patterns in microbiome data that conventional methods miss.
The 13th Heidelberg Laureate Forum discussed AI's impact on mathematics, with mathematicians debating its benefits and drawbacks. The forum also covered topics such as quantum threats, trustworthy computation, and the role of mathematics in society.
Researchers at the Stowers Institute used AlphaFold2 and evolutionary data to predict protein structures in aphids, which were previously inaccessible to AI. The study reveals a common architectural plan among 2,400 BICYCLE proteins, showcasing the evolution's role in helping AI predict protein structures.
Researchers developed a machine learning-based method to identify chemical compounds that can safely repel honey bees from pesticide-treated crops. The model identified 130 compounds with strong potential as bee repellents, which were tested in lab and field experiments, confirming their efficacy.
Researchers developed an AI model that uses routine chest CT scans to identify patients at risk of pneumonitis, a potentially life-threatening form of lung inflammation. The model, called CIPHER, achieved high predictive power and maintained strong performance despite differences in patient populations and imaging protocols.
The 2026 World Medical Innovation Forum brought together 250+ leaders to discuss emerging treatments, AI, and scientific breakthroughs. Mass General Brigham's Chief Innovation Officer highlighted the forum's goal of accelerating innovation to improve patient care.
A new framework IDEAI organizes embodied AI around explicit intent, combining perception, knowledge, planning, action and values. It provides a structured commitment that can be inspected, revised and verified, guiding action and ensuring safety.
Researchers identified shared and disease-specific molecular changes in human brain tissue, contributing to neurodegeneration, psychiatric illness, and cognitive decline. The collection establishes a foundational resource for understanding brain disorders, accelerating discovery and therapy development.
A UCF researcher and her team used AI to screen existing FDA-approved drugs for potential treatments of schwannoma tumors in children with Neurofibromatosis type 2. The research identified 10 promising candidates, with the most favorable targets stopping tumor cell growth by impacting different cellular mechanisms.
Researchers at Bar-Ilan University and Weizmann Institute successfully engineered immune cells to recognize and attack cancer and viral targets. The enhanced T cells demonstrated stronger immune responses and greater cancer-killing ability, improving survival in mice.
Researchers developed an AI framework that can represent complex three-dimensional hydraulic conductivity fields more efficiently and use monitoring data to improve predictions of PFOA movement in groundwater. The framework, VA-LSGAN, compressed complex fields into a smaller set of variables while preserving important spatial patterns.
The Goldilocks Program aims to reduce the timeline for new health technologies from 12-18 months to 90 days through continuous feedback and testing. Key findings include the potential to move critical care toward a predictive, precision-based field that identifies and corrects dangerous immune system dysregulation.
A quadruped robot, RAIBO2, completed a 42.195 km marathon without battery swap, capturing 4 hours and 19 minutes of real-world running data. The research team developed design principles for long-distance operation, focusing on system-level efficiency and reliability.
The use of AI in consular affairs is increasing, with automated systems screening passport photos and pre-sorting visa applications. However, concerns arise about human oversight, with experts warning that AI can shape what information officials see before making decisions that affect people's lives.
Researchers from the University of East London have developed a new method to predict how different chemicals behave, which could help identify the most promising battery materials. The method, published in the Journal of the American Chemical Society, uses X-ray photoelectron spectroscopy and computer modeling to predict the behavior ...
The proposed framework integrates knowledge graphs and large language models to diagnose space TWTAs, achieving high adoption rates and robust stability, with a 94.56% adoption rate and 95% accuracy in real-scenario testing.
The Research Highlighter-MatchMaker Project streamlines research collaboration and funding discovery using AI and natural language processing. It offers a searchable portal and automated email recommendations to facilitate connections between researchers and funding opportunities.
Brodsky is exploring the translation potential of Microgrid Optima, a platform to optimize energy storage operations and minimize peak power demand. The project aims to address the rising energy demand by 2030, partly due to growing AI computational requirements.
Researchers identify brain circuit that supports flexible behavior, but is hijacked by cocaine to drive repetitive actions. Activating or suppressing opposing neural pathways can either promote rigid behavior or interrupt cocaine-driven repetitive behavior.
A novel multimodal auditory assessment platform combines brain-activity measurements and pupil-response monitoring to evaluate hearing-related disorders. The platform has been implemented for data collection and has shown promise in identifying features associated with abnormal auditory sensitivity.
The AudioSight platform combines smartphone-based pupillometry, low-cost near-infrared hardware, and AI analysis to screen for and track hearing-related disorders. The platform generates risk scores for auditory dysfunctions using a BiLSTM-attention model, offering a potential solution for widespread hearing disorders.
The Salk Institute's $18 million Bezos Earth Fund grant will test whether deeper-rooted soybeans can store more carbon in soil and withstand drought and disease. The project aims to develop and test soybean plants with deeper, stronger roots using artificial intelligence, field trials, and soil carbon studies.
Professor Ruibang Luo has been awarded the 2026 APEC Science Prize for Innovation, Research and Education for his groundbreaking AI-powered tools in genomic analysis. His open-source tools have been downloaded over 10 million times worldwide and are being integrated into healthcare, agricultural, and medical biotechnology industries.
Professor Jay Siegel emphasizes the need for a human-centric AI economy to mitigate negative externalities and ensure human fulfilment. He draws parallels with historical breakthroughs in chemistry and petroleum, highlighting the importance of factoring societal costs into technology implementation models.
The third year of 'AI Adventures: When AI Meets Games with Tencent' concludes with a more diverse cohort of students from around the world, exploring AI tools with traditional Chinese cultural elements. The programme offers students a layered understanding of AI, placing them in direct contact with industry knowledge and academic persp...
Researchers develop a new approach for reconstructing graphs with incomplete information, handling both feature and structure completion. The EWS-RGCN model uses separate channels and a multi-level contrastive graph mask autoencoder to overcome limitations of weak supervision and limited labeled nodes.
The EMERGE project establishes a philosophical, mathematical and technological framework for collaborative awareness in artificial systems. Researchers found that people can understand an artificial system as aware without assuming subjective experience, and that increasing awareness can improve performance.
Researchers found that AI chatbots, including ChatGPT, produce less sophisticated responses when prompted with language associated with women, compared to male-coded prompts. The study highlights the potential for AI tools to perpetuate biases and disadvantage women in professional communication.
A KAUST-led team developed AI tool Unify to compare cell types across distant species, revealing similarities in biological meaning and improving human health research. Unify distinguished between identical genes, similar genes with evolved new jobs, and different genes with independently evolved functions.
New research reveals the rise of AI-generated deepfake physician scams and the potential of robotics in bedside nursing care. While robots can perform repetitive tasks, human involvement is still necessary due to the complexity and flexibility of real-life clinical settings. Additionally, deepfakes are used to exploit online users for ...
University of Tennessee MSE faculty members Sergei Kalinin and Mahshid Ahmadi contribute to the DOE's Genesis Mission, a $5 billion program developing AI tools for energy and scientific research. Kalinin's research accelerates materials discovery, while Ahmadi works on defending agentic AIs from adversarial attacks.
Researchers propose combining AI with environmental digital twins to create a framework for predicting risks, testing interventions, and supporting adaptive decisions. These systems could connect real-time observations, physical and data-driven models, AI algorithms, and decision feedback in a continuously evolving loop.
Researchers developed an AI-powered framework to optimize 2D material growth, enabling rapid process optimization and customized synthesis. The approach integrated machine learning and knowledge-driven reasoning to decipher multifactorial mechanisms.
Researchers at KAIST developed CURE, a technology that enables small AI models on smartphones to work efficiently with large models on servers, reducing server calls by an average of 55.61% while maintaining high accuracy. This approach allows for faster decisions with less server support.
The journal refines its evaluation criteria to emphasize translational impact, external validation, and actionable implementation in healthcare environments. Submissions will be evaluated based on real-world clinical applicability and systematic implementation of informatics solutions.
GrantsMate, an AI-driven research support platform, streamlines research workflows by integrating funding discovery, collaborator identification, and institutional policy guidance. The platform provides personalized recommendations and a conversational interface to simplify research administration and improve funding prospects.
SafeSeal embeds identifiable marks in LLM outputs while preserving content quality and robustness against removal attempts. It achieves this through named entity recognition and context-aware synonym replacement, ensuring high semantic similarity and entity preservation.
The Cross-Cultural Leadership Story Puzzle Game is an interactive, role-based platform that simulates authentic workplace challenges, encouraging cross-cultural communication and collaboration. AI-supported assessment provides personalized feedback, facilitating knowledge retention and improvement of leadership competencies.