Details

Autor(en) / Beteiligte

• Schoonhoven, Richard
• Skorikov, Alexander
• Palenstijn, Willem Jan
• Pelt, Daniël M
• Hendriksen, Allard A
• Batenburg, K Joost

Titel

How auto-differentiation can improve CT workflows: classical algorithms in a modern framework

Ist Teil von

• Optics express, 2024-03, Vol.32 (6), p.9019-9041

Ort / Verlag

United States

Erscheinungsjahr

2024

Quelle

EZB Electronic Journals Library

Beschreibungen/Notizen

• Many of the recent successes of deep learning-based approaches have been enabled by a framework of flexible, composable computational blocks with their parameters adjusted through an automatic differentiation mechanism to implement various data processing tasks. In this work, we explore how the same philosophy can be applied to existing "classical" (i.e., non-learning) algorithms, focusing on computed tomography (CT) as application field. We apply four key design principles of this approach for CT workflow design: end-to-end optimization, explicit quality criteria, declarative algorithm construction by building the forward model, and use of existing classical algorithms as computational blocks. Through four case studies, we demonstrate that auto-differentiation is remarkably effective beyond the boundaries of neural-network training, extending to CT workflows containing varied combinations of classical and machine learning algorithms.