Please use this identifier to cite or link to this item: https://doi.org/10.21256/zhaw-25181
Publication type: Conference poster
Type of review: Peer review (abstract)
Title: Deep learning-based simultaneous multi-phase deformable image registration of sparse 4D-CBCT
Authors: Herzig, Ivo
Paysan, Pascal
Scheib, Stefan
Züst, Alexander
Schilling, Frank-Peter
Montoya, Javier
Amirian, Mohammadreza
Stadelmann, Thilo
Eggenberger Hotz, Peter
Füchslin, Rudolf Marcel
Lichtensteiger, Lukas
et. al: No
DOI: 10.1002/mp.15769
10.21256/zhaw-25181
Published in: Medical Physics
Volume(Issue): 49
Issue: 6
Page(s): e325
Pages to: e326
Conference details: AAPM Annual Meeting, Washington, DC, USA, 10-14 July 2022
Issue Date: 9-Jun-2022
Publisher / Ed. Institution: American Association of Physicists in Medicine
Language: English
Subjects: Deep learning; Deformable image registration; CBCT; Medical imaging; Artificial intelligence
Subject (DDC): 006: Special computer methods
616: Internal medicine and diseases
Abstract: Purpose: Respiratory gated 4D-CBCT suffers from sparseness artefacts caused by the limited number of projections available for each respiratory phase/amplitude. These artefacts severely impact deformable image registration methods used to extract motion information. We use deep learning-based methods to predict displacement vector-fields (DVF) from sparse 4D-CBCT images to alleviate the impacts of sparseness artefacts. Methods: We trained U-Net-type convolutional neural network models to predict multiple (10) DVFs in a single forward pass given multiple sparse, gated CBCT and an optional artefact-free reference image as inputs. The predicted DVFs are used to warp the reference image to the different motion states, resulting in an artefact-free image for each state. The supervised training uses data generated by a motion simulation framework. The training dataset consists of 560 simulated 4D-CBCT images of 56 different patients; the generated data include fully sampled ground-truth images that are used to train the network. We compare the results of our method to pairwise image registration (reference image to single sparse image) using a) the deeds algorithm and b) VoxelMorph with image pair inputs. Results: We show that our method clearly outperforms pairwise registration using the deeds algorithm alone. PSNR improved from 25.8 to 46.4, SSIM from 0.9296 to 0.9999. In addition, the runtime of our learning-based method is orders of magnitude shorter (2 seconds instead of 10 minutes). Our results also indicate slightly improved performance compared to pairwise registration (delta-PSNR=1.2). We also trained a model that does not require the artefact-free reference image (which is usually not available) during inference demonstrating only marginally compromised results (delta-PSNR=-0.8). Conclusion: To the best of our knowledge, this is the first time CNNs are used to predict multi-phase DVFs in a single forward pass. This enables novel applications such as 4D-auto-segmentation, motion compensated image reconstruction, motion analyses, and patient motion modeling.
URI: https://digitalcollection.zhaw.ch/handle/11475/25181
Fulltext version: Published version
License (according to publishing contract): Licence according to publishing contract
Departement: School of Engineering
Organisational Unit: Centre for Artificial Intelligence (CAI)
Institute of Applied Mathematics and Physics (IAMP)
Published as part of the ZHAW project: DIR3CT: Deep Image Reconstruction through X-Ray Projection-based 3D Learning of Computed Tomography Volumes
Appears in collections:Publikationen School of Engineering

Files in This Item:
File Description SizeFormat 
2022_Herzig-etal_Sparse4DCBCT-multiphase-DIR_AAPM-eposter.pdf713.76 kBAdobe PDFThumbnail
View/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.