Integrating acquisition and AI in tomography

In recent years, deep learning has shown large potential in imaging applications, including tomography. At the same time, recent developments in hardware and software at experimental facilities enable unprecedented flexibility and possibilities in designing advanced acquisition schemes.

However, in many state-of-the-art applications of deep learning to imaging problems, the acquisition of data is performed using protocols that are not optimized for subsequent application of deep learning. In addition, the deep learning methods themselves are often trained and applied without exploiting knowledge about the way the data was acquired.

In this Lorentz Center workshop, we aim to bring together experts in tomographic data acquisition and experts in deep learning for scientific images to investigate novel approaches for integrating these two domains. The availability of powerful deep learning algorithms and the emergence of new capabilities of tomographic scanners provide a unique opportunity to create a next generation of AI-integrated tomography systems and workflows.

The deadline to register without an invitation was Sunday 24 September. Please note that the workshop is full. New applications will not be accepted.