Corrigendum to “Machine learning applied to HR-pQCT images improves fracture discrimination provided by DXA and clinical risk factors” [Bone. 2023 Mar:168:116653] (Bone (2023) 168, (S8756328222003301), (10.1016/j.bone.2022.116653))

Shengyu Lu, Nicholas R. Fuggle, Leo D. Westbury, Mícheál Ó Breasail, Gregorio Bevilacqua, Kate A. Ward, Elaine M. Dennison, Sasan Mahmoodi, Mahesan Niranjan, Cyrus Cooper*

*Corresponding author for this work

Research output: Contribution to journalComment/debate (Academic Journal)peer-review

Abstract

The authors regret the omission of the following italicised text within the methods section of the above paper: A sample of 22 consecutive slices was selected for each HR-pQCT image. There were fewer participants with previous fractures compared to those without. Therefore, an oversampling strategy was used for individuals with previous fractures [33], such that multiple samples were taken from the scans of those with previous fractures. This was performed assuming intra-scan homogeneity and to provide balance to the machine learning dataset. The authors would like to apologise for any inconvenience caused.

Original languageEnglish
Article number117071
JournalBone
Volume182
DOIs
Publication statusPublished - May 2024

Bibliographical note

Publisher Copyright:
© 2024 The Author(s)

Fingerprint

Dive into the research topics of 'Corrigendum to “Machine learning applied to HR-pQCT images improves fracture discrimination provided by DXA and clinical risk factors” [Bone. 2023 Mar:168:116653] (Bone (2023) 168, (S8756328222003301), (10.1016/j.bone.2022.116653))'. Together they form a unique fingerprint.

Cite this