An electron microscopy image can capture atoms arranged in a crystal lattice or defects threading through a semiconductor material, but turning that image into materials insight can take weeks of careful analysis. Now, an autonomous artificial intelligence platform developed at Cornell can do that work in minutes.
The EMSeek platform, reported April 1 in Science Advances, streamlines materials research by identifying key features in a microscopy image, determining the crystal structure, predicting material properties, comparing results with existing scientific literature, and generating a report within a single, integrated workflow.
"Electron microscopy produces incredibly rich information, but the bottleneck is often turning those images into usable scientific understanding," said corresponding author Fengqi You, the Roxanne E. and Michael J. Zak Professor in Energy Systems at the Cornell Duffield College of Engineering. You is also co-director of the Cornell University AI for Science Institute.
To read more, click here.