Skip navigation

Supplementary Information for Data-driven image mechanics (D2IM): a deep learning approach to predict displacement and strain fields from undeformed X-ray tomography images - Evaluation of bone mechanics

Supplementary Information for Data-driven image mechanics (D2IM): a deep learning approach to predict displacement and strain fields from undeformed X-ray tomography images - Evaluation of bone mechanics

Soar, Peter ORCID: 0000-0003-1745-9443 , Palanca, Marco, Enrico, Dall'Ara and Tozzi, Gianluca (2024) Supplementary Information for Data-driven image mechanics (D2IM): a deep learning approach to predict displacement and strain fields from undeformed X-ray tomography images - Evaluation of bone mechanics. [Dataset] (Unpublished)

[img] Plain Text (README for the dataset)
README.txt - Supplemental Material

Download (514B)
[img] Archive (ZIP) (Contains jpg images of example U Displacement predictions for D2IM test dataset)
testU.zip - Supplemental Material
Available under License Creative Commons Attribution.

Download (25MB)
[img] Archive (ZIP) (Contains jpg images of example V Displacement predictions for D2IM test dataset)
testV.zip - Supplemental Material
Available under License Creative Commons Attribution.

Download (24MB)
[img] Archive (ZIP) (Contains jpg images of example W Displacement predictions for D2IM test dataset)
testW.zip - Supplemental Material
Available under License Creative Commons Attribution.

Download (24MB)

Abstract

In the paper a focus was made on a few cases of w displacement predictions deemed as both interesting and of indicative general behaviour by the authors. However, D2IM predicts full fields for all three displacement components (u, v and w), with 26 predictions being made in the test dataset. Due to the improved performance on predictions of w (for reasons explained in section 3 of the paper), ultimately this was the field focussed on and no results of the other displacement components were presented or discussed in the paper. However, these other fields still make overall good predictions of the behaviour, so for completeness’s sake the entire set of predictions have been provided so they can be examined and appraised by any interested parties. In this supplemental file, links are provided to Greenwich Universities file repository GALA, where the full set of 26 predictions for all three displacement components can be downloaded. These images each contain summarised in a single image – the input, the ground truth displacement field, the predicted displacement field and the absolute prediction error.

Item Type: Dataset
Uncontrolled Keywords: D2IM; bone; X-ray computed tomography; digital volume correlation; deep learning; convolutional neural network
Subjects: Q Science > Q Science (General)
Q Science > QA Mathematics
Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Faculty / School / Research Centre / Research Group: Faculty of Engineering & Science
Faculty of Engineering & Science > School of Computing & Mathematical Sciences (CMS)
Faculty of Engineering & Science > School of Engineering (ENG)
Last Modified: 10 Oct 2024 08:53
URI: http://gala.gre.ac.uk/id/eprint/46147

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year

View more statistics