How deep learning is accelerating multiscale design of porous electrodes for flow cells
Peer-Reviewed Publication
This month, we’re focusing on artificial intelligence (AI), a topic that continues to capture attention everywhere. Here, you’ll find the latest research news, insights, and discoveries shaping how AI is being developed and used across the world.
Updates every hour. Last Updated: 2-Jan-2026 07:11 ET (2-Jan-2026 12:11 GMT/UTC)
Researchers from The Hong Kong University of Science and Technology and the Southern University of Science and Technology have developed a novel deep learning neural network, Electrode Net. By introducing signed distance fields and three-dimensional convolutional neural networks, this method can significantly accelerate electrode design while maintaining high accuracy. It is widely applicable to fuel cells, water electrolyzers, flow batteries, etc.
USC researchers found a potential way to extend the clock on post-stroke treatment and enable better stroke recovery - an experimental stem cell therapy to help repair damaged brain tissue, co-developed by scientists at the Keck School of Medicine of USC, the University of Zurich and ETH Zurich in Switzerland. A study in the journal Nature Communications showed that a stem cell transplant performed one week after an ischemic stroke in mice led to recovery. Rust and his colleagues reprogrammed human blood cells into neural stem cells — which can mature into neurons — and transplanted them into the damaged brain tissue of mice that had strokes. After five weeks, the researchers compared their recovery to a group of mice from the same litter that had strokes but underwent surgery without transplantation. The brains of the mice that received transplanted neural stem cells showed more robust signs of recovery than those of untreated mice. The transplant recipients’ brains had less inflammation, more growth of neurons and blood vessels, and more connectivity among neurons than the brains of the mice that did not receive transplanted cells. The treated mice also had less leakage from the blood-brain barrier, which is important for normal brain function and acts as a filter to keep harmful substances out of the brain. To measure function, the researchers used artificial intelligence to closely track the movement of the animals’ limbs while walking and climbing up a ladder with irregular rungs. The team found that treated mice fully recovered the fine motor skills tested in the climbing task five weeks after the transplants. By the end of the study, their gait also improved significantly compared to mice that received a sham surgery.
If AI’s intrinsic risks are real, governmental regulation and ethical frameworks are unlikely to contain them. Drawing on social theory, it highlights myths about the state’s capacity, global enforcement challenges, rapid technological decentralization, and the ambiguity of moral norms. The author presents a skeptical view that “meaning well” does not ensure effective outcomes, cautioning against overreliance on governments and ethics to mitigate advanced AI risks.
The Cardiovascular Research Foundation® (CRF®) and Jon DeHaan Foundation today announced the launch of the TCT AI Lab, a groundbreaking new program debuting at TCT® 2025, October 25–28 at the Moscone Center in San Francisco. The TCT AI Lab is a first-of-its-kind destination dedicated to integrating artificial intelligence into clinical practice. Over three intensive days, clinicians will progress from the fundamentals of AI to hands-on clinical applications – guided by leading innovators at the forefront of digital medicine.
Simon Fraser University’s Burnaby campus is once again home to Canada’s most powerful academic supercomputer, following the installation of a new system, named Fir. The new Fir system replaces the Cedar supercomputer, housed at the Cedar Supercomputing Centre (the Centre) at SFU. Fir is ranked number 78 in the TOP500 list of the world’s most powerful supercomputers, the only Canadian system in the top 100 worldwide.
“The new Fir supercomputer represents a much needed, major upgrade to the national Canadian computing infrastructure,” says Dugan O’Neil, SFU’s Vice President Research and Innovation. “The growing importance of data for research in all types of academia and in industry continues to drive demand for high-performance computing, and we are meeting that need. Fir will drive research across Canada and I have no doubt it will facilitate significant advances in a wide range of fields in the coming years.”