Researchers at UW–Madison have come up with a proof of concept for using electrochemistry to extract lithium from spent lithium-iron-phosphate (LFP) batteries, which have been widely adopted by major EV manufacturers like Tesla and China’s BYD.

Lower back pain is the most common musculoskeletal issue in the U.S. and a top cause of global disability. To tackle this, researchers have developed a groundbreaking AI-powered system that automates patient-specific lumbar spine modeling. By merging deep learning with biomechanical simulation, the new method slashes model prep time by nearly 98% – from more than 24 hours to just 30 minutes – while preserving clinical accuracy. This innovation enables faster, more consistent diagnoses and personalized treatment planning.

University of Massachusetts Amherst researchers and scientists at Embr Labs, a Boston-based start-up, have developed an AI-driven algorithm that can accurately predict nearly 70% of hot flashes before they’re perceived. The work, featured in the journal Psychophysiology, will be incorporated into the Embr Wave, a wearable wrist device clinically proven to manage hot flashes.

A new artificial intelligence model found previously undetected signals in routine heart tests that strongly predict which patients will suffer potentially deadly complications after surgery. The model significantly outperformed risk scores currently relied upon by doctors.

The federally-funded work by Johns Hopkins University researchers, which turns standard and inexpensive test results into a potentially life-saving tool, could transform decision-making and risk calculation for both patients and surgeons.

Crucial drugs such as crucial drugs such as statins, antibiotics and HIV inhibitors are among some of the most expensive because they require a chiral center ― a property in which a molecule cannot be superimposed with its mirror image. In a new study, University of Maine researchers explore a new, cost-reducing pathway to produce one of the key building blocks for chiral drugs, (S)-3-hydroxy-γ-butyrolactone (HBL), from glucose at high concentrations and yields. Because glucose can be derived from any lignocellulosic feedstock, this process opens a new door for the sustainable production of HBL. 

Genes are the building blocks of life, and the genetic code provides the instructions for the complex processes that make organisms function. But how and why did it come to be the way it is? A recent study from the University of Illinois Urbana-Champaign sheds new light on the origin and evolution of the genetic code, providing valuable insights for genetic engineering and bioinformatics.

Researchers at Lehigh University have received a $2 million grant from the National Science Foundation to explore how biological computation could make artificial intelligence (AI) more energy-efficient and powerful. The team, led by Yevgeny Berdichevsky, associate professor of bioengineering and electrical and computer engineering, will study neurons within engineered brain organoids—three-dimensional structures grown from adult stem cells that resemble developing brain tissue. By arranging and stimulating neurons in controlled ways, the researchers aim to replicate how the brain processes dynamic information, such as moving images. The interdisciplinary project combines neuroscience, bioengineering, electrical engineering, and computer science to create a proof of concept for biological computation and to inspire new AI algorithms. The researchers will also assess the ethical, social, and legal considerations of using brain organoids in computation.

Studying miniature analogs of natural earthquakes in the lab, MIT geologists quantified how much energy from the quake goes into heat, shaking, and fracturing. The research could help seismologists predict the likelihood of quakes in seismically active regions.