Traditional geotechnical investigations provide data only at discrete borehole locations, leaving vast areas uncharacterized. This spatial gap often leads to unforeseen ground conditions during construction, causing costly delays, design modifications, and occasionally catastrophic failures. Now, a novel integrated geophysical-machine learning approach, using k-means clustering technique, by a team of researchers from Shibaura Institute of Technology provides continuous subsurface characterization, enabling evidence-based decision-making throughout project lifecycles.

Understanding molecular diversity is fundamental to biomedical research and diagnostics, but existing analytical tools struggle to distinguish subtle variations in the structure or composition among biomolecules. A new method using solid-state nanopores — tiny tunnels that a protein or other molecule can pass through — and machine learning suggests a way to help identify and classify proteins more accurately.

This study presents a sustainable strategy for upcycling ocean-derived CO₂ and opens new avenues for electrochemically driven biochemical synthesis.

The public are invited to share their most vibrant memories as researchers seek ways to help us recall experiences more vividly

Miniaturized electronics and intricate objects require a certain finesse. Here, researchers look into the development of a machine capable of these fine movements, as well as identifying and categorizing the objects using machine learning.

Researchers at the University of Edinburgh have discovered that childhood cognitive function shares significant genetic overlap with longevity, showing a genetic correlation of 0.35. Using genome-wide association data from over 400,000 participants, the study provides the first evidence linking childhood intelligence genes with lifespan. The findings, published in Genomic Psychiatry, suggest that genetic variants associated with higher childhood cognitive abilities are also linked to longer parental lifespans.