Texas A&M University Associate Professor Dr. Jinsil Hwaryoung Seo utilizes immersive technology and artificial intelligence to create virtual hands-on learning experiences that enhance student engagement and skill development.

Advances in technology have led to the miniaturization of many mechanical, electronic, chemical and biomedical products, and with that, an evolution in the way these tiny components and parts are transported is necessary to follow. Transport systems, such as those based on conveyor belts, suffer from the challenge of friction, which drastically slows the speed and precision of small transport. Researchers from YOKOHAMA National University addressed this issue by developing an untethered levitation device capable of moving in all directions. The frictionless design allows for ultrafast, agile movement that can prove to be very valuable in machine assembly, biomedical and chemical applications via contactless transport.

Researchers published their results in Advanced Intelligent Systems in July 2025.

Not all blood vessels play equal roles in brain health. To uncover how the brain regulates its complex circulation, researchers built a detailed computer model of mouse brain vasculature. Focusing on transitional zone vessels—crucial links between arteries and capillaries—the model simulates how each segment adjusts like a tiny valve. It reveals how the brain stabilizes blood flow during pressure shifts or heightened activity, offering new insights into brain protection and potential breakthroughs in diagnosing stroke, Alzheimer’s, and traumatic brain injuries.

In a new study, scientists at the University of Missouri looked deep into the universe and found something unexpected. Using infrared images taken from NASA’s powerful James Webb Space Telescope (JWST), they identified 300 objects that were brighter than they should be.

Researchers from Shenyang University of Chemical Technology explored the application potential of the Catal-GPT in catalyst design, which was built upon the qwen2 large language model, proposing a new paradigm for AI-driven catalyst development. The results showed that the qwen2 model could provide the complete preparation workflows and the detailed optimization suggestions through conversational interaction. Future work aims for cross-system adaptability to transform catalyst discovery from trial-and-error to precision targeting.