When people watch a scene in the film Jurassic Park where a giant dinosaur walks toward them, they naturally imagine a heavy, rumbling sound, as if the ground were shaking. This is because humans predict sound by considering not only the shape of an object, but also physical properties such as its size, weight, and speed of movement. However, existing video-to-audio generation AI mainly generates sound based on the category of objects or scene information in the video, and has not sufficiently reflected physical properties that vary depending on weight or speed.

Detecting melanoma before it becomes visible is a major challenge in dermatology. Now, with researchers from Université de Montréal, scientists at Université du Québec’s Institut national de la recherche scientifique (INRS) have developed a promising solution tested on mice.

Called SMEAR-ULM, it’s a high-tech system that can detect skin cancers at their earliest stages by measuring tiny temperature variations at the surface of the skin. Led by INRS professor Jinyang Liang, the research team’s findings are published in Nature Sensors. 

The work was carried out in close collaboration with several research teams, including ones led by INRS professor Fiorenzo Vetrone and, at UdeM, pharmacology professor Davide Brambilla and medical professor Sylvain Meloche. 

The potential impact of the work is significant, the scientists say. 

Student teams from Monta Vista High School in Cupertino, California, and Sierra Vista Middle School in Irvine, California, took top honors at the 42nd Annual Science Olympiad National Tournament at the University of Southern California on Saturday, May 23, 2026. This is the second time that Monta Vista High School has captured the title and adds a back-to-back championship for Sierra Vista Middle School. 

Anyone seeking to purchase healthy products at the supermarket needs clear guidance. Nutritional labelling schemes, such as Nutri-Score labels, are designed to help with that. The Nutri-Score front-of-pack label shows a five-colour system from favourable nutritional composition in green (A) through to unfavourable in red (E). The scheme is optional on products across the EU but has been adopted by companies in several countries, for instance, Germany and the Netherlands. Some supermarkets also print a grey version on the price tag, but the colourful version on the shelf also attracts more attention here. Researchers at Göttingen University – in collaboration with Osnabrück University of Applied Sciences, the German Institute of Food Technology and Giessen University – studied consumers’ visual attention on these label formats. Their research showed colourful labelling catches the eye more frequently and for longer than monochrome. Particularly frequent and prolonged glances at the dark green A or the yellow C were usually accompanied by choosing this particular product. The red E, on the other hand, did not have the desired warning effect: even when this unfavourable rating was viewed frequently, the products still ended up in shopping baskets. The study was published in the journal Food Quality and Preference.

Rigoberto Advincula, a polymer researcher at the Department of Energy’s Oak Ridge National Laboratory and a Governor’s Chair Professor at the University of Tennessee, Knoxville, has received the American Filtration and Separations Society’s Frank Tiller Award at FILTCON26 in Pittsburgh. The honor recognizes his lifetime leadership and scientific impact in fluid-particle separation, citing advances in new materials, 3D-printed membranes, and smart separation surfaces. A highly decorated fellow of multiple scientific societies, Advincula has authored 513 publications and holds 15 issued patents, with additional patents pending.

West Virginia University is launching the WVU Rare Earth Elements Initiative, or WVU REE. The initiative unites researchers who focus on rare earths and critical minerals as they explore a growing range of U.S. feedstocks. In tandem, WVU has established the for-profit startup, Mission Critical Materials, to commercialize 10 years of the University’s proven AMDREE technology.

A new AI model developed by UC San Diego researchers could make it possible to predict treatment response based only on a tumor's genetics.

(Toronto, May 25, 2026) JMIR Publications released a story on how 3D-printed anatomical models are being used to improve surgical outcomes in its News and Perspectives section.