An AI-powered tool called MELD Graph is revolutionizing epilepsy care by detecting subtle brain abnormalities that radiologists often miss.
By analyzing global MRI data, the tool improves diagnosis speed, increases access to surgical treatment, and cuts healthcare costs. Though not yet in clinical use, it is already helping doctors identify operable lesions, offering hope to epilepsy patients worldwide.
AI Breakthrough in Epilepsy Detection.
There’s an arms race in medicine—scientists design drugs to treat lethal bacterial infections, but bacteria can evolve defenses to those drugs, sending the researchers back to square one. In an article published in the Journal of the American Chemical Society, a University of California, Irvine-led team describes the development of a drug candidate that can stop bacteria before they have a chance to cause harm.
“The issue with antibiotics is this crisis of antibiotic resistance,” said Sophia Padilla, a Ph.D. candidate in chemistry and lead author of the new study. “When it comes to antibiotics, bacteria can evolve defenses against them—they’re becoming stronger and always getting better at protecting themselves.”
About 35,000 people in the U.S. die each year from antibiotic-resistant bacterial infections from pathogens like Staphylococcus, while about 2.8 million people suffer from bacteria-related illnesses.
Firefly Aerospace’s Blue Ghost spacecraft took some unique photos of our home planet on its Mission 1 flight to the moon in late January and early February. The spacecraft is now in orbit around the moon, with a planned landing date of March 2.
Scientists have taken a huge step forward in perfecting the method to synthesize lonsdaleite.
Scientists at Penn State have harnessed a unique property called incipient ferroelectricity to create a new type of computer memory that could revolutionize how electronic devices work, such as using much less energy and operating in extreme environments like outer space.
They published their work, which focuses on multifunctional two-dimensional field-effect transistors (FETs), in Nature Communications. FETs are advanced electronic devices that use ultra-thin layers of materials to control electrical signals, offering multiple functions like switching, sensing or memory in a compact form.
They are ferroelectric-like, meaning the direction of their electric conduction can be reversed when an external electric field is applied to the system. FETs are essential in computing, since the ferroelectric-like property allows them to shift signals.
Silicon is the best-known semiconductor material. However, controlled nanostructuring drastically alters the material’s properties. Using a specially developed etching apparatus, a team at HZB has now produced mesoporous silicon layers with countless tiny pores and investigated their electrical and thermal conductivity.
For the first time, the researchers elucidated the electronic transport mechanism in this mesoporous silicon. The material has great potential for applications and could also be used to thermally insulate qubits for quantum computers. The work is published in Small Structures.
Mesoporous silicon is crystalline silicon with disordered nanometer-sized pores. The material has a huge internal surface area and is also biocompatible. This opens up a wide range of potential applications, from biosensors to battery anodes and capacitors. In addition, the material’s exceptionally low thermal conductivity suggests applications as thermal insulator.
Researchers at the University of Bayreuth present novel electrospun nonwovens in Science Advances that exhibit an unusual combination of high electrical conductivity and extremely low thermal conductivity.
The nonwovens represent a breakthrough in materials research: it has been possible to decouple electrical and thermal conductivity based on a simple-to-implement material concept. The nonwovens are made of carbon and silicon-based ceramic via electrospinning process and are attractive for technological applications, for example, in energy technology and electronics. They can be manufactured and processed cost-effectively on an industrial scale.
Normally, high electrical conductivity is associated with high thermal conductivity, and low thermal conductivity goes with low electrical conductivity. However, in many high-tech industries, there is growing interest in multifunctional materials that that combine good electric with low thermal transport.
Researchers at the Technical University of Munich (TUM) have invented an entirely new field of microscopy called nuclear spin microscopy. The team can visualize magnetic signals of nuclear magnetic resonance with a microscope. Quantum sensors convert the signals into light, enabling extremely high-resolution optical imaging.
Magnetic resonance imaging (MRI) scanners are known for their ability to look deep into the human body and create images of organs and tissues. The new method, published in the journal Nature Communications, extends this technique to the realm of microscopic detail.
“The quantum sensors used make it possible to convert magnetic resonance signals into optical signals. These signals are captured by a camera and displayed as images,” explains Dominik Bucher, Professor of Quantum Sensing and researcher at the Cluster of Excellence Munich Center for Quantum Science and Technology (MCQST).
Researchers have developed a battery that can convert nuclear energy into electricity via light emission, a new study suggests.
Nuclear power plants, which generate about 20% of all electricity produced in the United States, produce almost no greenhouse gas emissions. However, these systems do create radioactive waste, which can be dangerous to human health and the environment. Safely disposing of this waste can be challenging.
Using a combination of scintillator crystals, high-density materials that emit light when they absorb radiation, and solar cells, the team, led by researchers from The Ohio State University, demonstrated that ambient gamma radiation could be harvested to produce a strong enough electric output to power microelectronics, like microchips.
A team of Carnegie Mellon University researchers set out to see how accurately large language models (LLMs) can match the style of text written by humans. Their findings were recently published in the Proceedings of the National Academy of Sciences.
“We humans, we adapt how we write and how we speak to the situation. Sometimes we’re formal or informal, or there are different styles for different contexts,” said Alex Reinhart, lead author and associate teaching professor in the Department of Statistics & Data Science.
“What we learned is that LLMs, like ChatGPT and Llama, write a certain way, and they don’t necessarily adapt to the writing style. The context and their style are actually very distinctive from how humans normally write or speak in different contexts. Nobody has measured or quantified this in the way we were able to do.”