Task-based Generation of Optimized Projection Sets using Differentiable Ranking

Schneider LS, Thies M, Syben C, Schielein R, Unberath M, Maier A (2023)

Publication Language: English

Publication Type: Conference contribution, Conference Contribution

Publication year: 2023

Conference Proceedings Title: Fully3D 2023 Proceedings

Event location: Stony Brook State University of New York US

URI: https://arxiv.org/abs/2303.11724

DOI: 10.48550/arXiv.2303.11724

Abstract

We present a method for selecting valuable projections in computed tomography (CT) scans to enhance image reconstruction and diagnosis. The approach integrates two important factors, projection-based detectability and data completeness, into a single feed-forward neural network. The network evaluates the value of projections, processes them through a differentiable ranking function and makes the final selection using a straight-through estimator. Data completeness is ensured through the label provided during training. The approach eliminates the need for heuristically enforcing data completeness, which may exclude valuable projections. The method is evaluated on simulated data in a non-destructive testing scenario, where the aim is to maximize the reconstruction quality within a specified region of interest. We achieve comparable results to previous methods, laying the foundation for using reconstruction-based loss functions to learn the selection of projections.

Authors with CRIS profile

Linda-Sophie Schneider Lehrstuhl für Informatik 5 (Mustererkennung) Mareike Thies Lehrstuhl für Informatik 5 (Mustererkennung) Andreas Maier Lehrstuhl für Informatik 5 (Mustererkennung)

Involved external institutions

Fraunhofer Development Center X-ray Technology (EZRT) DE Germany (DE) Johns Hopkins Hospital US United States (USA) (US)

How to cite

APA:

Schneider, L.-S., Thies, M., Syben, C., Schielein, R., Unberath, M., & Maier, A. (2023). Task-based Generation of Optimized Projection Sets using Differentiable Ranking. In Fully3D 2023 Proceedings. Stony Brook State University of New York, US.

MLA:

Schneider, Linda-Sophie, et al. "Task-based Generation of Optimized Projection Sets using Differentiable Ranking." Proceedings of the Fully3D 2023, Stony Brook State University of New York 2023.

BibTeX: Download