Get Full Government Meeting Transcripts, Videos, & Alerts Forever!
Get email alerts on the Ai Imaging topic
No spam. Unsubscribe anytime.
Argonne official demonstrates AI reconstruction that turns detector patterns into real‑time nanoscale images
Summary
Laurent Chapon of Argonne National Laboratory showed how AI can convert complex detector patterns into real‑time 3D images—reducing analysis that once took days to near real time and operating at about 200 images per second on scattering facilities.
Laurent Chapon, associate lab director for photo sciences at Argonne National Laboratory, demonstrated how AI models reconstruct X‑ray scattering detector patterns into interpretable three‑dimensional images in near real time.
Chapon showed a microchip example and said the laboratory collective measures the equivalent of "150,000,000 movies every year." He described building vision models from large data banks that enable real‑time analysis and anomaly detection at approximately 200 images per second, turning pipelines that once took days or weeks into near real time.
Chapon called the transformation "unbelievable" and said the capability will affect manufacturing, materials discovery and biotech applications by enabling machine‑actionable representations of experimental results.
