Lifeboat Foundation: Safeguarding Humanity
Toggle light / dark theme

Blog

Get the latest international news and world events from around the world.

Log in for authorized contributors

Treatment with norrin helps maintain healthy blood vessels and preserve vision in diabetic rats

Millions of people in the U.S. are affected by diabetic retinopathy, a complication of diabetes that affects the eyes. The condition is caused by damage to the blood vessels in the retina, the light-sensitive tissue at the back of the eye. At first, diabetic retinopathy might not cause any symptoms, or the problems might be mild. However, if the underlying diabetes remains poorly managed, it can lead to blindness.

In a new study, University of Michigan researchers showed that treatment with the molecule norrin can prevent vision loss in diabetic rats. The study is published in the journal Diabetes.

Diabetic retinopathy treatment Current therapies for diabetic retinopathy target vascular endothelial growth factor, or VEGF, a protein that causes new blood vessels to form.

Robots gain ‘muscle memory’ for planning their next moves in milliseconds

If “to err is human,” you could say robots are superhuman. They’re more error-prone than we are, even when tasked with things like home and factory chores, which we might consider easy. One reason for these mistakes is that robots struggle to extract relevant details from the past and act on them in real time, even with vision-language models (VLMs), which interpret visual inputs such as videos and images along with natural language.

Recent approaches use AI agents, which are semiautonomous programs based on VLMs, that, in this case, decide when robots should interpret specific details from a long history of camera frames. However, this can lead to errors, high compute costs and poor reactivity.

A new approach called “Workspace Models” from researchers at MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) and Carnegie Mellon University’s Machine Learning Department (MLD) reformulates the agentic pipeline used to compress history for robots.

Ultrafast 3D imaging captures fleeting motion at 150,000 frames per second

A research team from the Institut national de la recherche scientifique (INRS), led by professor Jinyang Liang, is pushing the limits of ultrafast imaging with a new technology capable of capturing three-dimensional images at speeds of up to 150,000 frames per second. Published in the journal Nature Communications, this study establishes a new world record in profilometry.

The breakthrough could transform the way scientists observe and measure extremely fast, hard-to-reproduce phenomena. The technology, called C-BLIP (chromatically band-limited illumination profilometry), makes it possible to visualize collisions, aerodynamic motion and ultrasonic vibrations in three dimensions. It could find applications in fields ranging from mechanics and aerodynamics to microfluidics, high-rate material testing and additive manufacturing.

Using C-BLIP, the researchers successfully captured and measured a variety of transient phenomena, including the collision between a projectile and a turbofan blade, the aerodynamics of a falling maple samara and the vibrations of a loudspeaker membrane.

Inner Workings: A massive star dies without a bang, revealing the sensitive nature of supernovae

In 2008, a huge red star in another galaxy reached the end of its life. A star as heavy as this one, born with 25 times the mass of the Sun, was supposed to go out in a fiery flash of light known as a supernova, millions or billions of times brighter than our Sun. But this one refused to play the role of drama queen. Instead, it brightened just a little, then vanished, possibly leaving behind a black hole.

No one had ever seen one of these huge red stars wink out of existence with so little fuss before. It was a sign that the lives and deaths of these stars are more complex than our simplest theories had claimed. “As amazing and important and fun and exciting as this is, it’s not a surprise,” says Stan Woosley at the University of California, Santa Cruz. In fact, the discovery may help explain why the massive stars in computer models often fail to blow up.

We’ve got biology wrong, and that’s why the Human Genome Project failed

The Human Genome Project promised genetic-based cures for all the major diseases of mankind, from cancer, to Alzheimer, to schizophrenia. A quarter of a century later, and these sweeping cures haven’t materialized. Biologist Denis Noble and his co-authors Raymond Noble and Reine Bourret argue that this failure reveals a flawed foundation in modern biology: treating genes as the system’s controllers rather than its tools. For 60 years this central dogma has mislead biology. Once we overthrow it, we can transform our understanding of how species evolve.

Figure 1: diagram of the Central Dogma

Biology during the twentieth century was firmly based on Watson and Crick’s famous work on the structure of DNA. They discovered (with important data input from Wilkins and Franklin) that DNA forms double helical threads, each thread being a complementary sequence to the other. The nucleotide bases C and G join together, while A and T also do so. So, a sequence like AACGTTCC in one thread is matched by the complement TTGCAAGG. They realized that this has an important implication. When the threads are unwound, each thread can attract its complementary sequence automatically, so creating two new threads. It is easy to imagine their excitement since the quantum mechanics pioneer, Erwin Schrödinger, had proposed self-replication “like a crystal” in his book What is Life? a decade or so earlier. Crick incorporated all of this into his Central Dogma of Molecular Biology.

Physicists Say They’ve Sidestepped a Nearly 100-Year-Old Quantum Limit Using Hindsight

The Heisenberg uncertainty principle, introduced by German physicist Werner Heisenberg in 1927, puts forward the concept of a certain fuzziness in physics. It says that if one characteristic of a quantum particle can be predicted with precision, a paired characteristic becomes less certain.

Position and momentum are the most often-mentioned of these pairs, known as non-commuting variables. Being able to ascertain where a quantum particle is makes it harder to judge its movement, and vice versa.

That fuzziness makes it more difficult to take accurate measurements of weaker signals in physics – in everything from the LIGO gravitational wave detector to the atomic clocks that underpin systems like GPS.