Learn how to break bad habits using neuroscience, not willpower. Science-backed strategies that rewire cravings and stick.
Get the latest international news and world events from around the world.
Teaching robots to map large environments
A robot searching for workers trapped in a partially collapsed mine shaft must rapidly generate a map of the scene and identify its location within that scene as it navigates the treacherous terrain.
Researchers have recently started building powerful machine-learning models to perform this complex task using only images from the robot’s onboard cameras, but even the best models can only process a few images at a time. In a real-world disaster where every second counts, a search-and-rescue robot would need to quickly traverse large areas and process thousands of images to complete its mission.
To overcome this problem, MIT researchers drew on ideas from both recent artificial intelligence vision models and classical computer vision to develop a new system that can process an arbitrary number of images. Their system accurately generates 3D maps of complicated scenes like a crowded office corridor in a matter of seconds.
Combination therapy for Burkitt’s lymphoma dramatically improves cure rates in mice
Burkitt’s lymphoma is a rare and aggressive blood cancer characterized by a translocation of the MYC gene. It occurs most often in children and young adults. In recent years, CAR-T cell therapy—often referred to as a “living drug” and administered as a single dose—has been approved for certain types of blood cancer, offering hope for a cure even in severe cases. However, its effectiveness against Burkitt’s lymphoma has been limited. Moreover, developing drugs that directly target MYC—the root cause of this cancer—has proven challenging for decades.
Recently, a study led by Dr. Hiroshi Kotani, Assistant Professor at Kanazawa University in collaboration with a scientist at Roswell Park Comprehensive Cancer Center in Buffalo, NY, U.S., revealed that a SUMOylation inhibitor can suppress MYC activity. Building on this finding, the research team investigated whether combining CAR-T therapy with the SUMOylation inhibitor TAK-981 could improve outcomes for Burkitt’s lymphoma. The research is published in Signal Transduction and Targeted Therapy.
The team first confirmed that the SUMOylation inhibitor effectively slowed the growth of Burkitt’s lymphoma cells and altered their signaling pathways. They then examined the inhibitor’s effect on CAR-T cells and discovered a dual role: While it initially activated the CAR-T cells in a way that could hinder long-term effectiveness, it also triggered a built-in “safety brake” mechanism. These insights suggested that using only a limited dose of the inhibitor could maximize the benefits of CAR-T therapy as a durable, living treatment.
Robots That Feel Pain: Scientists Develop Human-Like Artificial Skin With Instant Reflexes
This technology is quite different from the workings of most robots used today. Many robots lack the ability to sense touch at all, and those that do can usually only detect simple pressure. Such robots lack self-protective reflexes.
In these systems, touch information first travels to the software, where it is analysed step-by-step before a response is determined. This process might be acceptable for robots working within safety enclosures in factories, but it’s insufficient for humanoid robots working in close proximity to humans.
Unlike humans, robots cannot heal themselves. However, scientists say the best alternative is quick and easy repair. According to them, the new skin converts touch signals into neural-like pulses and activates protective reflexes upon detecting pain. The skin can also detect damage, and thanks to its modular design, damaged sections can be quickly replaced.
Alternative wireless technology achieves stable outdoor data transmission
The approach addresses key challenges in visible light communication, including pulse distortion and sunlight interference.
Scientists have developed a low-cost visible light communication (VLC) system using commercially available hardware that enables stable data transmission even under strong ambient light.
The team achieved reliable outdoor VLC at data rates of up to 3.48 Mbit/s over distances of several meters by implementing a newly designed 8B13B coding scheme on an FPGA and interfacing it with a Raspberry Pi.
The approach addresses key challenges in VLC, including pulse distortion and sunlight interference, and offers a practical path toward intelligent transportation system (ITS) applications.
CRISPR screen uncovers hundreds of genes required for brain development
Which genes are required for turning embryonic stem cells into brain cells, and what happens when this process goes wrong? In a new study published today in Nature Neuroscience, researchers led by Prof. Sagiv Shifman from The Institute of Life Sciences at The Hebrew University of Jerusalem, in collaboration with Prof. Binnaz Yalcin from INSERM, France, used genome-wide CRISPR knockout screens to identify genes that are needed for early brain development.
The study set out to answer a straightforward question: which genes are required for the proper development of brain cells?
Using CRISPR-based gene-editing methods, the researchers systematically and individually “switched off” roughly 20,000 genes to study their role in brain development. They performed the screen in embryonic stem cells while the cells changed into brain cells. By disrupting genes one by one, the team could see which genes are required for this transition to proceed normally.
