Optimising inter-patient image registration for image-based data mining in breast radiotherapy
Issued Date
2024-10-01
Resource Type
eISSN
24056316
Scopus ID
2-s2.0-85203409326
Journal Title
Physics and Imaging in Radiation Oncology
Volume
32
Rights Holder(s)
SCOPUS
Bibliographic Citation
Physics and Imaging in Radiation Oncology Vol.32 (2024)
Suggested Citation
Jaikuna T., Wilson F., Azria D., Chang-Claude J., De Santis M.C., Gutiérrez-Enríquez S., van Herk M., Hoskin P., Kotzki L., Lambrecht M., Lingard Z., Seibold P., Seoane A., Sperk E., Paul Symonds R., Talbot C.J., Rancati T., Rattay T., Reyes V., Rosenstein B.S., de Ruysscher D., Vega A., Veldeman L., Webb A., West C.M., Aznar M.C., Vasquez Osorio E. Optimising inter-patient image registration for image-based data mining in breast radiotherapy. Physics and Imaging in Radiation Oncology Vol.32 (2024). doi:10.1016/j.phro.2024.100635 Retrieved from: https://repository.li.mahidol.ac.th/handle/20.500.14594/101218
Title
Optimising inter-patient image registration for image-based data mining in breast radiotherapy
Author(s)
Jaikuna T.
Wilson F.
Azria D.
Chang-Claude J.
De Santis M.C.
Gutiérrez-Enríquez S.
van Herk M.
Hoskin P.
Kotzki L.
Lambrecht M.
Lingard Z.
Seibold P.
Seoane A.
Sperk E.
Paul Symonds R.
Talbot C.J.
Rancati T.
Rattay T.
Reyes V.
Rosenstein B.S.
de Ruysscher D.
Vega A.
Veldeman L.
Webb A.
West C.M.
Aznar M.C.
Vasquez Osorio E.
Wilson F.
Azria D.
Chang-Claude J.
De Santis M.C.
Gutiérrez-Enríquez S.
van Herk M.
Hoskin P.
Kotzki L.
Lambrecht M.
Lingard Z.
Seibold P.
Seoane A.
Sperk E.
Paul Symonds R.
Talbot C.J.
Rancati T.
Rattay T.
Reyes V.
Rosenstein B.S.
de Ruysscher D.
Vega A.
Veldeman L.
Webb A.
West C.M.
Aznar M.C.
Vasquez Osorio E.
Author's Affiliation
Instituto de Investigación Sanitaria de Santiago de Compostela
Medizinische Fakultät Mannheim
Siriraj Hospital
College of Life Sciences
Vall d‘Hebron Institut de Oncologia
Institut de Recherche en Cancérologie de Montpellier
Centro de Investigación Biomédica en Red de Enfermedades Raras
Universitair Ziekenhuis Gent
University of Leicester
KU Leuven
German Cancer Research Center
School of Medical Sciences
Hôpital Universitaire Carémeau
Icahn School of Medicine at Mount Sinai
Hospital Universitari Vall d'Hebron
Maastricht Universitair Medisch Centrum+
Fondazione IRCCS Istituto Nazionale dei Tumori, Milan
Universitätsklinikum Hamburg-Eppendorf
Fondazione IRCCS Isituto Nazionale dei Tumori
Grupo de Medicina Xenómica (USC)
Medizinische Fakultät Mannheim
Siriraj Hospital
College of Life Sciences
Vall d‘Hebron Institut de Oncologia
Institut de Recherche en Cancérologie de Montpellier
Centro de Investigación Biomédica en Red de Enfermedades Raras
Universitair Ziekenhuis Gent
University of Leicester
KU Leuven
German Cancer Research Center
School of Medical Sciences
Hôpital Universitaire Carémeau
Icahn School of Medicine at Mount Sinai
Hospital Universitari Vall d'Hebron
Maastricht Universitair Medisch Centrum+
Fondazione IRCCS Istituto Nazionale dei Tumori, Milan
Universitätsklinikum Hamburg-Eppendorf
Fondazione IRCCS Isituto Nazionale dei Tumori
Grupo de Medicina Xenómica (USC)
Corresponding Author(s)
Other Contributor(s)
Abstract
Background and purpose: Image-based data mining (IBDM) requires spatial normalisation to reference anatomy, which is challenging in breast radiotherapy due to variations in the treatment position, breast shape and volume. We aim to optimise spatial normalisation for breast IBDM. Materials and methods: Data from 996 patients treated with radiotherapy for early-stage breast cancer, recruited in the REQUITE study, were included. Patients were treated supine (n = 811), with either bilateral or ipsilateral arm(s) raised (551/260, respectively) or in prone position (n = 185). Four deformable image registration (DIR) configurations for extrathoracic spatial normalisation were tested. We selected the best-performing DIR configuration and further investigated two pathways: i) registering prone/supine cohorts independently and ii) registering all patients to a supine reference. The impact of arm positioning in the supine cohort was quantified. DIR accuracy was estimated using Normalised Cross Correlation (NCC), Dice Similarity Coefficient (DSC), mean Distance to Agreement (MDA), 95 % Hausdorff Distance (95 %HD), and inter-patient landmark registration uncertainty (ILRU). Results: DIR using B-spline and normalised mutual information (NMI) performed the best across all evaluation metrics. Supine-supine registrations yielded highest accuracy (0.98 ± 0.01, 0.91 ± 0.04, 0.23 ± 0.19 cm, 1.17 ± 1.18 cm, 0.51 ± 0.26 cm for NCC, DSC, MDA, 95 %HD, and ILRU), followed by prone-prone and supine-prone registrations. Arm positioning had no significant impact on registration performance. For the best DIR strategy, uncertainty of 0.44 and 0.81 cm in the breast and shoulder regions was found. Conclusions: B-spline algorithm using NMI and registered supine and prone cohorts independently provides the most optimal spatial normalisation strategy for breast IBDM.