A new machine, the first of its kind in Europe, allows users to visualise what happens when tiny particles hit a surface three times faster than a bullet. This information will enable engineers to make better materials and coatings for aircraft parts, medical implants, and for simulating intergalactic collisions.

This study presents a machine learning approach to estimate the top-of-atmosphere (TOA) shortwave radiation flux from Deep Space Climate Observatory / Earth Polychromatic Imaging Camera (DSCOVR/EPIC) satellite observations. 

A novel deep learning–based framework that dramatically improves the accuracy of forecasts, even when data are limited. The task involved using past atmospheric variables to predict five key surface weather indicators—including temperature, wind, and precipitation—every 6 hours for the next 5 days.

A new international study led by Kaunas University of Technology reveals that digital mental health start-ups reaching unicorn status tend to secure private funding early, prioritise simple, scalable services, and use subscription models –  while reliance on state funding may actually hinder growth. The research, based on 125 ventures (including 12 unicorns), suggests success stems from strategic, complementary choices rather than technological complexity alone. Researchers call for a shift in European investment policy to focus more on market relevance and user impact rather than deep tech or rigid scientific criteria.

A research team at Shanghai Jiao Tong University has developed a systematic framework that combines semantic-segmentation AI models with experimental validation to analyze catalyst layers (CLs). Using silver nanoparticles as catalysts and Nafion ionomer as a binder, the team fabricated CLs with varying ionomer-to-catalyst (I/C) ratios (0.2, 0.4, and 0.6) to dissect how composition affects performance.