Janse, Markus H. A. https://orcid.org/0000-0002-0604-1204
Janssen, Liselore M.
Wolters-van der Ben, Elian J. M.
Moman, Maaike R.
Viergever, Max A.
van Diest, Paul J.
Gilhuijs, Kenneth G. A.
Funding for this research was provided by:
Horizon 2020 (755333)
Article History
Received: 10 March 2025
Revised: 26 May 2025
Accepted: 5 June 2025
First Online: 6 August 2025
Compliance with ethical standards
:
: The scientific guarantor of this publication is Dr. K.G.A. Gilhuijs.
: The authors of this manuscript declare no relationships with any companies, whose products or services may be related to the subject matter of the article.
: Dr. B.B.L. Penning de Vries kindly provided statistical advice for this manuscript, Dr. K.G.A. Gilhuijs has significant statistical expertise.
: Written informed consent was waived by the Institutional Review Board.
: Institutional Review Board approval was obtained from METC Utrecht (no. 19-245 and no. 19-396).
: Study subjects in the training and testing cohorts have been previously reported on. All subjects have previously been described and used in Janse et al [ ]; the current study reuses a segmentation network described for a new purpose (RCB prediction). The test set was acquired as part of a prospective trial. The results of this trial were previously reported in Janssen et al [ ]. That study mainly concerns itself with the relation between liquid biopsies and classical MRI features, related to Residual Cancer Burden (RCB), and does not use advanced radiomic techniques. Finally, both sets were previously reported on in a study investigating the relation between tumor-infiltrating lymphocytes (TILs) and MRI in relation to disease prognosis [ ]. The current study also used the TILs measurement for Supplementary Fig. . All studies had minor variations in inclusion/exclusion criteria for their purposes, thus there are slight changes in overlap.
: