Center for Biomolecular Structure Lecture Series

" AI Tools in X-ray Computed Tomography: Applications, Drawbacks, and Key Insights"

Presented by Julio-Cesar daSilva, CNRS Neel Institute, France

Wednesday, August 20, 2025, 1:30 pm — Videoconference / Virtual Event (see link below)

I will present the most commonly used AI tools in X-ray computed tomography, highlighting their potential drawbacks. Moreover, I will discuss the basic machine learning concepts essential for effectively utilizing these tools and their implementation at the Hub AI for Tomography, AIAX, located in Grenoble, France. After examining densely connected neural networks, convolutional neural networks, and encoder-decoder architectures, I will provide a brief explanation of how the U-Net, MSDNet, Noise2Noise, TomoGAN, and GANRec networks function within the field of tomography. Their primary applications focus on image denoising and 3D image segmentation. The examples provided will be sourced from hard X-ray tomography applied to both biological and materials science specimens. The key takeaway from this presentation is that while AI-generated results can be advantageous, they need to be critically evaluated, as they always require human involvement for refinement.

Videoconference Instructions

Registration Required

Join Videoconference More Information

22058  |  INT/EXT  |  Events Calendar

 

Not all computers/devices will add this event to your calendar automatically.

A calendar event file named "calendar.ics" will be placed in your downloads location. Depending on how your device/computer is configured, you may have to locate this file and double click on it to add the event to your calendar.

Event dates, times, and locations are subject to change. Event details will not be updated automatically once you add this event to your own calendar. Check the Lab's Events Calendar to ensure that you have the latest event information.