D-net: Siamese based network with mutual attention for volume alignment

Conference item

Alignment of contrast and non contrast-enhanced imaging is essential for quantification of changes in several biomedical applications. In particular, the extraction of cartilage shape from contrast-enhanced Computed Tomography (CT) of tibiae requires accurate alignment of the bone, currently performed manually. Existing deep learning-based methods for alignment require a common template or are limited in rotation range. Therefore, we present a novel network, D-net, to estimate arbitrary rotation and translation between 3D CT scans that additionally does not require a prior template. D-net is an extension to the branched Siamese encoder-decoder structure connected by new mutual, non-local links, which efficiently capture long-range connections of similar features between two branches. The 3D supervised network is trained and validated using preclinical CT scans of mouse tibiae with and without contrast enhancement in cartilage. The presented results show a significant improvement in the estimation of CT alignment, outperforming the current comparable methods.

Authors: Zheng, J-Q; Lim, NH; Papież, BW

• Publication status: Published
• Peer review status: Peer reviewed
• Publisher: Springer
• Host title: Shape in Medical Imaging. ShapeMI 2020
• Pages: 73-84
• Series: Lecture Notes in Computer Science
• Series number: 12474
• Place of publication: Cham, Switzerland
• Publication date: 2020-10-30
• Event title: ShapeMI MICCAI 2020: Workshop on Shape in Medical Imaging
• Event location: Lima, Peru
• Event start date: 2020-10-04
• Event end date: 2020-10-04
• DOI: 10.1007/978-3-030-61056-2_6
• EISBN: 9783030610562
• ISBN: 9783030610555
• Language: English
• Keywords: deep learning; image registration; Siamese encoder decoder; mutual non-local link; FFR