Science AI

MemBrain v2 automates membrane analysis in cryo-ET images

MemBrain v2 automates membrane analysis in cryo-ET images Image: Primary
Researchers at Helmholtz Munich, the Technical University of Munich and the University of Basel released MemBrain v2, an open-source AI tool for analyzing membranes in cryo-electron-tomography images. The software combines membrane detection, protein localization and spatial analysis, replacing work the researchers say previously took weeks with processing that can take hours. In one test, researchers annotated protein-complex positions on one membrane and reported a 91% F1 score when locating them on additional membranes. The study was published in Nature Methods.
Sources
Phys.org
Published by Tech & Business, a media brand covering technology and business. This story was sourced from Phys.org and reviewed by the T&B editorial agent team.
Back to Newswire
Keep reading
Full wire
AI Science
AI Science

ORNL demonstrates AI-guided targeting for atomic-force microscopy

Researchers at Oak Ridge National Laboratory developed SimuScan, an AI framework for atomic-force microscopy that uses synthetic training images to identify nanoscale features and direct follow-up scans. In demonstrations involvi...

Phys.org
Science AI
Science AI

Physics-informed AI extracts brain fluid flow speeds from MRI data

Researchers at the University of Rochester, Brown University and the University of Copenhagen reported in Science Advances a physics-informed AI method that estimates fluid flow velocity and tissue permeability from MRI data, a me...

SciTechDaily
Science AI
Science AI

BenchShield preprint targets reward hacking in AI-agent benchmarks

A new preprint describes BenchShield, an instrumentation layer for detecting reward hacking in interactive LLM-agent benchmarks. The system uses a finite lifecycle model of reward-relevant events, combining static taint analysis b...

cs.AI updates on arXiv.org