Research on making sense of scientific and medical data. Covers AI error-correction, single-cell search engines, optical computing, and measurement reliability.
A new framework trains AI models to screen AI-generated content using performance data from past marketing campaigns. The models provide content recommendations and ratings, streamlining the decision-making process for marketers. Human capital plays a vital role in the successful use of these new technologies.
Researchers develop standardized measures to evaluate mobile crisis programs' impact on police response, revealing higher diversion rates than previously estimated. The framework provides communities with a consistent way to measure their own programs' effectiveness.
A targeted quality improvement campaign by University of Cincinnati emergency medicine physicians safely reduced unnecessary head and cervical spine CT scans in low-risk trauma patients, saving patients $6.2 million in healthcare costs and avoiding significant radiation exposure. The campaign, which included continuing education sessio...
Researchers at Pennington Biomedical developed GVC-Calc to analyze continuous glucose monitoring data, providing a scalable and transparent approach. The platform supports batch processing, integrates individual-level data with cohort-level results, and offers consensus clinical metrics alongside exploratory measures.
Large language models can support environmental research through systematic literature screening, relational knowledge mining, and quantitative data extraction. These approaches can transform unstructured scientific literature into structured information that can be reused in databases, models, risk assessments, and environmental decis...
A study by Politecnico di Milano and the University of Insubria tested THInkPen, a sensorized pen, on over 700 children to assess writing difficulties. The results revealed significant relationships between digital indicators and clinical scores, confirming the ability of the pen to reliably reflect performance characteristics.
The World Trade Center Health Program has provided critical insights into the health effects of 9/11, revealing the long-term consequences of exposure to toxic substances. The program's data has informed care delivery and clinical understanding of the affected population.
KAIST researchers develop SafeQL, a technology that identifies and selectively corrects errors in AI-generated SQL queries, reducing the need to regenerate entire queries. The technology improves data retrieval accuracy and speed, accelerating the adoption of AI work assistants in enterprise environments.
Researchers at the Max Delbrück Center have developed a search engine, Malva, to analyze single-cell RNA data, enabling rapid analysis of millions of cells worldwide. Malva simplifies the task of wading through data from thousands of experiments and provides insights into RNA biology, cancer, and disease mechanisms.
Optical convolution computation enables parallel light propagation and multiplexing for faster and more energy-efficient computing systems. The review organizes the field into two paradigms: definition-based and theorem-based, which leverage mathematical principles to implement convolution operations in the physical domain.
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 randomized clinical trial found spectacle lenses with highly aspherical lenslets reduced myopia progression and axial elongation in children over 24 months. The study supports the use of these lenses as a safe and effective myopia control intervention.
A cohort study of pregnant females with opioid use disorder found higher discontinuation rates for Black females compared to White females, with less disparity for methadone treatment. Persistent racial and ethnic inequities in treatment retention were observed during and after pregnancy.
A decade-long review of Kids First DRC shows how shared data expands research opportunities across diseases, institutions, and scientific disciplines. The resource has supported findings with potential to improve diagnosis, risk assessment, and treatment in pediatric care.
A post hoc analysis of a cluster randomized clinical trial found that the Real-time Prescription Benefit Tool (RTPB) increased fill rates for high-cost medications, especially among patients from low-income communities. However, the tool's recommendations were only made for a small proportion of orders, limiting the study's applicability.
Researchers reviewed grant data from 2015-2022, supporting investment in a strong evidence base for obesity and cancer research. Key areas of focus include obesity's impact on cancer treatment and survivorship, as well as disparities in cancer outcomes.
The KAIST research team developed an explainable AI technology that detects patterns of foreign-linked influence operations in online news comments. The model identifies 23,998 accounts exhibiting patterns consistent with public-opinion manipulation, targeting division and confrontation within Korean society.
A hybrid lifestyle program showed potential benefits for cardiometabolic health among adherent participants with obesity, but failed to achieve significant weight loss. The study suggests further investigation into the program's effects on metabolic health.
The new integrations provide enterprise AI agents direct access to live, structured research data from Dimensions' 430M+ interconnected records. This allows for AI-assisted analytics across one of the world's most comprehensive linked views of global research activity.
A cohort study finds that childhood history of maltreatment is associated with increased biological age in late midlife, potentially impacting health outcomes. The study's epigenetic clock analysis sheds light on the long-term effects of early life trauma on aging processes.
A new study suggests that artificial intelligence can enhance the implementation of Medicaid work requirements by helping state agencies keep eligible individuals enrolled. AI tools can analyze existing databases to verify compliance or exemption status, reducing documentation difficulties and administrative complexities. However, huma...
This study investigates the link between vegan diets and dietary energy density, finding that adopting a vegan diet may lead to lower energy intake. The randomized clinical trial analyzed data from over 10,000 participants, suggesting that a well-planned vegan diet can be an effective way to reduce energy consumption.
A new Cochrane review analyzed 11 studies involving 2,524 women undergoing embryo transfer, finding no reliable evidence that preparatory techniques improve IVF success rates. The techniques examined included straightening the utero-cervical angle and removing cervical mucus before transfer.
A new AI model developed by Japan Advanced Institute of Science and Technology identifies the most relevant information in videos, significantly reducing analysis time. The model achieved state-of-the-art performance while using only 15.42% of available visual features and reduced processing time per video by approximately 65%.
A cohort study analyzing factors associated with abandoned buprenorphine prescriptions found that US adults are increasingly dropping their medication for opioid use disorder. The study, published in JAMA Network Open, examines the decline of prescribed buprenorphine from 2020 to 2024.
Physicists have developed a method to visualize three-dimensional wavefunctions of molecules, enabling the study of molecular interactions. The technique, which uses a table-top soft-X-ray laser and powerful computer algorithms, allows for the imaging of features smaller than atomic scales.
Researchers used connected speech analysis to distinguish between nonfluent, logopenic and semantic PPA variants with high accuracy. The study also showed expected atrophy associations and discriminated common neuropathologic classes in an autopsy-confirmed subset.
The EU project B-Cubed developed automated data pipelines to standardize biodiversity data, creating 'species occurrence cubes' that support consistent indicator reporting. These data cubes enable the derivation of meaningful measures of biodiversity status and trends, allowing for timely responses to dynamic threats.
A cross-sectional study identified critical methodological limitations in RCTs of CHM, including unclear hypotheses and inadequate allocation concealment. The study highlights the need for improved planning and conduct of CHM RCTs, with superior designs published in English-language journals.
The University of Chicago Data Science Institute and Financial Mathematics Program launch a new industry-aligned certificate in quantitative development, combining training in advanced software engineering with financial expertise. Students will gain experience building real-world systems and preparing them to succeed in leading firms.
A nationally representative analysis found substantial gaps in LDL-C control, with increasing disparities across primary prevention categories and underuse of lipid-lowering therapy. The study provides a contemporary population-level benchmark for the updated guideline framework.
A new study uses machine learning to predict chemical toxicity in rare and endangered species, reducing the need for direct biological testing. The model achieved strong performance predicting acute and chronic toxicity, with life stage being a key factor.
The Data Sciences Institute at the University of Toronto has been awarded $1 million in Claude API credits to support AI-enabled research. Researchers will gain access to cutting-edge AI tools, enabling discovery, analysis, and innovation across disciplines.
A new statistical test developed by Stanford researchers assesses whether personalizing interventions is worth it. The K-fold personalization test (KPT) provides an estimate of expected benefits and a range for potential gains, helping users understand trade-offs involved in tailoring interventions to individuals. This tool can aid inf...
A new study proposes a switching-based pervasive augmented reality framework that integrates location-based AR, deep learning, and context-awareness to improve landmark recognition. The framework significantly enhanced detection accuracy compared with conventional LBAR systems, while demonstrating high user satisfaction.
A meta-analysis of individual participant data found that amyloid PET quantitation using centiloids can accurately diagnose Alzheimer's disease. The study suggests that scans in the 11-26 Centiloid range require caution, and a positivity cutoff of around 18 Centiloids is recommended.
Researchers developed a photospike-based TRNG that harnesses unpredictable light-induced electrical charges to generate true random numbers. The device passed all 15 randomness tests and remained stable over millions of cycles, making it suitable for image authentication and deepfake detection.
Researchers from Kyoto University and ETH Zurich analyzed large earthquake sequences in Japan, finding that b-values are controlled more by mainshock location than time-dependent changes. The study suggests local geological conditions play a key role in shaping earthquake size distributions.
A new catalog centralizes publicly available soil moisture content datasets, streamlining data discovery and comparison. The UB-SMDC portal provides a unified platform for researchers to efficiently access and analyze global soil moisture data.
A study by King's College London found that students with declining school performance are at a higher risk of contact with the criminal justice system. Researchers identified 'signals' in school records indicating pupils who may need support, highlighting opportunities for early intervention.
A new study estimates that by 2100, only 38% of the world's population will live in large cities, with 450 million fewer people projected to reside in cities over a century than current trends suggest. Urban growth slows as countries urbanize, with smaller cities growing faster than larger ones initially.
Digital twins are expanding rapidly, using real-time data to simulate and analyze systems before applying them in the real world. The study emphasizes the need for interoperable architectures, machine-readable metadata, and standardized trust frameworks to address challenges such as privacy, cybersecurity, and uncertainty.
Climate conditions are increasingly limiting direct air free cooling in data centers, particularly across the tropics and southeastern US. Projections indicate continued expansion of these constraints with sustained warming and increasing humidity.
Researchers developed a tandem neural network that rapidly infers key semiconductor material properties from simple transistor measurements, outperforming conventional approaches. The system produces results in under one millisecond with near-perfect accuracy.
Rates of new obesity cases rose by 4% between 2019 and 2025, with the largest increases seen among young adults. The study also found significant disparities in obesity rates across sociodemographic groups and geographical regions, highlighting the need for deep-seated change to address the underlying factors driving obesity.
A recent study found that the FDA approval of semaglutide for weight management led to a significant increase in calls to poison control centers, with over 8,000 cases reported by 2023. The majority of errors were preventable and stemmed from unintentional dosing or therapeutic mistakes.
The project aims to systematically map how individual pairs of cells influence each other, with the goal of understanding cell-cell communication in health and disease. By characterizing the cellular dyad, scientists can determine which cell influenced which partner, when the interaction began and what changed as a result.
New research from the University of Kansas finds that AI technology can be used to deliver relevant ads without collecting personal data. The studies show that AI systems can generate relevance by interpreting content structure, semantic cues and emotional tone within a webpage.
Researchers have found that cats show comparable patterns of brain deterioration as they age, similar to humans. This study, part of the Translating Time project, aims to unlock new insights into human ageing and disease by comparing cat and human brain development.
Researchers developed a new lidar system that simultaneously measures distance, velocity and surface material properties in a scene. The system uses polarization information to extract this data with high precision and accuracy.
A recent study analyzing 5.8 billion medicines dispensed in England from 2019 to 2024 found significant health disparities, with nearly twice as many medicines prescribed to the most deprived groups as those in the least deprived groups by age 40.
The study uses smartwatch data to record how match events affect fans' heart rate and stress levels. Fans from Eastern and Southern Europe and Turkey are underrepresented, so the research team aims to include as many participants as possible.
A study led by UC Riverside highlights the importance of rigorous participant verification procedures in online health research to maintain data quality. Researchers found that combining automated screening, human review, and participant verification can substantially improve confidence in online research findings.
Researchers propose using low-dose lithium orotate to slow dementia progression in mild cognitive impairment, leveraging established neuroprotective mechanisms. The approach aims to mitigate kidney and thyroid risks associated with higher-dose formulations, offering a potentially accessible and inexpensive treatment option.
Researchers developed FLIM Playground to streamline FLIM data analysis, extracting lifetime information from cells and visualizing patterns. The platform reduces the need for multiple software tools and quality checks, increasing efficiency and reproducibility.
TurboLynx, developed by POSTECH researchers, analyzes complex, interconnected data up to 184 times faster than existing systems. The engine groups similar data together and processes them collectively, reducing unnecessary memory usage and enabling efficient analytical queries.
A study by Rice University researchers found that hitting with a slower swing speed can lead to more contact, but also sacrifices power. The study analyzed MLB swing-tracking data and found that batters who adjust their approach with two strikes do not necessarily get better results.
In an exploratory analysis, finerenone was found to slow kidney function decline, reduce albuminuria, and lower the risk of kidney failure or substantial loss of kidney function in patients with glomerular diseases. These findings suggest a crucial role for finerenone in preserving kidney function in this population.
A new curriculum at Iowa State University aims to prepare future math teachers to teach data science by leveraging their existing mathematical knowledge. The five-week module introduces students to the relationship between math and data science, using familiar structures such as algebra, geometry, and calculus to introduce new concepts.
A difference-in-differences analysis found that Georgia's Pathways to Coverage program worsened mental health among low-income adults. This suggests that work requirements may create barriers to coverage and care access, affecting population health and equity.