Fraunhofer MEVIS Image Registration Solutions for the Learn2Reg 2021 Challenge

Alessa Hering*, Annkristin Lange, Stefan Heldmann, Stephanie Häger, Sven Kuckertz

*Korrespondierende/r Autor/-in für diese Arbeit

Abstract

In this paper, we present our contribution to the learn2reg challenge. We applied the Fraunhofer MEVIS registration library RegLib comprehensively to all 3 tasks of the challenge, where we used a classic iterative registration method with NGF distance measure, second order curvature regularizer and a multi-level optimization scheme. We show that with our proposed method robust results can be achieved throughout all tasks resulting in the fourth place overall task and the best accuracy on the lung CT registration task.
OriginalspracheEnglisch
TitelLecture Notes in Computer Science
BandLNIP, volume 13166
Erscheinungsdatum2022
PublikationsstatusVeröffentlicht - 2022

Fingerprint

Untersuchen Sie die Forschungsthemen von „Fraunhofer MEVIS Image Registration Solutions for the Learn2Reg 2021 Challenge“. Zusammen bilden sie einen einzigartigen Fingerprint.

Zitieren