Tech & AI

Researchers have developed a way to estimate real-time traffic conditions using video from ordinary vehicles, eliminating the need for expensive infrastructure. The technique could cut deployment costs for cities and transport agencies while providing higher-resolution traffic data for congestion management and route optimization.EN

Researchers have designed a control system that lets industrial hydrogen electrolyzers act as emergency buffers for power grids overwhelmed by wind and solar production. The approach could let utilities build more renewable capacity while avoiding costly blackouts—and turn steel mills and chemical plants into paying grid operators.EN

Researchers have engineered a plastic-like material that conducts lithium ions reliably across a wide temperature range, potentially unlocking safer solid-state batteries for electric vehicles and grid storage. The breakthrough demonstrates how precise material architecture—layering different polymers at the nanoscale—can overcome a persistent bottleneck in battery technology.EN

Researchers have modeled a system where battery-equipped charging vehicles deliver power to autonomous electric vehicles while both are moving, potentially eliminating the need for expensive fixed charging infrastructure. The approach could reduce required battery capacity significantly while cutting travel times, offering a major cost advantage for fleet operators and cities deploying autonomous vehicle networks.EN

Researchers have achieved a 20.12% efficiency rating in organic solar cells by precisely engineering molecular coatings that improve how light converts to electricity. The breakthrough, which involved testing five different chemical variations, could accelerate commercial deployment of lightweight, flexible solar technology that costs less to manufacture than traditional silicon panels.EN

Swedish researchers have digitized and analyzed a unique archive of Italian language learners' speech recordings from 2001–2009, revealing patterns in how multilingual students develop grammar and vocabulary. The work offers insights for language education programs and shows how linguistic background influences learning outcomes—data increasingly valuable as companies expand international workforce training.EN

Researchers tested an AI-powered language learning system that automatically generates exercises from real texts, addressing a persistent challenge in heritage language education: teaching students with vastly different skill levels in the same classroom. The findings suggest intelligent tutoring systems can scale personalized learning without requiring teachers to create custom materials for each student.EN

Researchers found that 3D-printed dental implant models deliver haptic feedback—the tactile sensation of resistance and contact—comparable to real tissue, making them viable for surgeon training. The finding could reduce reliance on cadavers and live patients for initial surgical practice, lowering training costs while improving patient safety.EN

Researchers compared knowledge-based and data-driven methods for detecting bearing defects before catastrophic failure, finding both effective at identifying warning signals in vibration patterns. For manufacturers relying on rotating machinery, catching these early indicators could prevent costly downtime and equipment damage while extending asset lifespan.EN

Researchers have created the first comprehensive guide matching thermal storage systems with power conversion technologies for renewable energy plants. The findings could cut millions in capital costs for grid operators and renewable energy developers deciding how to store and dispatch power from intermittent sources like solar and wind.EN

A new clinical approach paper on reninoma—a rare hormone-secreting tumor—provides physicians with a structured diagnostic pathway for a condition often misdiagnosed as common hypertension. The framework could improve patient outcomes and reduce unnecessary treatments, relevant to healthcare systems managing rare disease detection efficiency.EN

Researchers have finally proved a decades-old conjecture about how networks shrink when connections are removed—a finding with implications for optimizing everything from supply chains to telecommunications infrastructure. The proof resolves a question that's been open since 1983 and could improve algorithms used to analyze complex systems.EN

A new field experiment validates computer vision systems for measuring where people look during behavioral research, potentially cutting costs and expanding the scale of psychology studies. The work matters because eye-tracking has long been expensive and lab-dependent—automating it could unlock insights for UX design, advertising, and market research.EN

A new simulation study reveals that survival analysis—a statistical method critical for predicting patient outcomes, product reliability, and treatment efficacy—can produce misleading results when researchers fail to account for multiple time-scales in their data. For companies and institutions relying on these analyses to make clinical, regulatory, or business decisions, the finding signals a potential source of systematic error that could affect millions of predictions.EN

Researchers have developed a continuous flow system that extracts useful compounds from spruce bark faster and more efficiently than existing methods, while using less solvent and energy. The breakthrough could make bark—currently considered waste in forestry—a profitable feedstock for chemical makers and help wood processors reduce costs.EN

Researchers have released Breeze 2.0, a free web-based platform that lets scientists quickly visualize and compare how different drugs affect cells. For pharmaceutical companies and research institutions, the tool could cut weeks off early-stage drug screening by automating what was previously manual, error-prone analysis work.EN

Researchers analyzed patterns in sick leave among working-age pedestrians injured in accidents, revealing how different types of injuries lead to varying absence lengths. The findings could help employers and insurers better predict workforce disruption and plan return-to-work programs more effectively.EN

Researchers have developed a deep learning algorithm that improves abdominal CT imaging clarity while maintaining diagnostic accuracy with lower radiation doses. The advance could reduce healthcare costs and patient risk across millions of annual scans, while helping hospitals compete on safety and image quality.EN

Researchers created high-resolution synthetic mammograms to retrain AI cancer detection systems, reducing performance gaps across demographic groups. The approach addresses a critical liability: medical AI systems that miss tumors in certain populations expose healthcare providers to litigation while widening care disparities.EN

Researchers identified which young children with pneumonia face the highest mortality risk using basic clinical data, even without pulse oximetry devices. The finding offers hospitals in low-income regions a practical way to triage patients and allocate scarce medical resources more effectively.EN

Scientists have created a detailed map of how experimental drugs interact with human liver and colorectal tissue using organoid cultures. The work could accelerate drug development by predicting safety and efficacy earlier in testing, potentially reducing costly late-stage failures and speeding medicines to market.EN

Researchers have updated how doctors and health systems should interpret evidence when studies disagree, a critical issue affecting clinical decisions affecting millions. The new framework clarifies when inconsistencies matter and when they don't—potentially saving time and money in medical decision-making while improving patient outcomes.EN

Researchers compared two laboratory growth systems for testing cancer drugs against patient tumor cells, a critical step toward tailored treatment plans. The findings could help pharmaceutical companies and hospitals faster identify which drugs work best for individual patients, potentially reducing trial-and-error treatment cycles.EN

Researchers demonstrated that bone can form outside the skeleton by packing calcium phosphate granules into a collagen pouch—a finding with significant implications for reconstructive surgery and dental implants. The approach could eventually reduce patient recovery times and expand treatment options for bone loss without invasive grafting procedures.EN

Researchers demonstrated that a dual-layer detector cone-beam CT system can rapidly assess stroke patients, potentially improving outcomes when minutes matter. The technology could reshape how hospitals prioritize and treat stroke cases, reducing delays that currently cost patients brain tissue and disability.EN