The importance of planning CT-based imaging features for machine learning-based prediction of pain response.

Autor: Llorián-Salvador Ó; Department of Radiation Oncology, Klinikum Rechts der Isar, Technical University of Munich (TUM), Ismaninger Straße 22, 81675, Munich, Germany.; Department for Bioinformatics and Computational Biology, Informatik 12, Technical University of Munich (TUM), Boltzmannstraße 3, 85748, Garching, Germany.; Institute of Organismic and Molecular Evolution, Johannes Gutenberg University Mainz, Hanns-Dieter-Hüsch-Weg 15, 55128, Mainz, Germany., Akhgar J; Department of Radiation Oncology, Klinikum Rechts der Isar, Technical University of Munich (TUM), Ismaninger Straße 22, 81675, Munich, Germany., Pigorsch S; Department of Radiation Oncology, Klinikum Rechts der Isar, Technical University of Munich (TUM), Ismaninger Straße 22, 81675, Munich, Germany., Borm K; Department of Radiation Oncology, Klinikum Rechts der Isar, Technical University of Munich (TUM), Ismaninger Straße 22, 81675, Munich, Germany., Münch S; Department of Radiation Oncology, Klinikum Rechts der Isar, Technical University of Munich (TUM), Ismaninger Straße 22, 81675, Munich, Germany., Bernhardt D; Department of Radiation Oncology, Klinikum Rechts der Isar, Technical University of Munich (TUM), Ismaninger Straße 22, 81675, Munich, Germany.; Department of Radiation Sciences (DRS), Institute of Radiation Medicine (IRM), Helmholtz Zentrum, 85764, München, Germany.; Deutsches Konsortium für Translationale Krebsforschung (DKTK), Partner Site Munich, 69120, Heidelberg, Germany., Rost B; Department for Bioinformatics and Computational Biology, Informatik 12, Technical University of Munich (TUM), Boltzmannstraße 3, 85748, Garching, Germany., Andrade-Navarro MA; Institute of Organismic and Molecular Evolution, Johannes Gutenberg University Mainz, Hanns-Dieter-Hüsch-Weg 15, 55128, Mainz, Germany., Combs SE; Department of Radiation Oncology, Klinikum Rechts der Isar, Technical University of Munich (TUM), Ismaninger Straße 22, 81675, Munich, Germany.; Department of Radiation Sciences (DRS), Institute of Radiation Medicine (IRM), Helmholtz Zentrum, 85764, München, Germany.; Deutsches Konsortium für Translationale Krebsforschung (DKTK), Partner Site Munich, 69120, Heidelberg, Germany., Peeken JC; Department of Radiation Oncology, Klinikum Rechts der Isar, Technical University of Munich (TUM), Ismaninger Straße 22, 81675, Munich, Germany. jan.peeken@tum.de.; Department of Radiation Sciences (DRS), Institute of Radiation Medicine (IRM), Helmholtz Zentrum, 85764, München, Germany. jan.peeken@tum.de.; Deutsches Konsortium für Translationale Krebsforschung (DKTK), Partner Site Munich, 69120, Heidelberg, Germany. jan.peeken@tum.de.
Jazyk: angličtina
Zdroj: Scientific reports [Sci Rep] 2023 Oct 13; Vol. 13 (1), pp. 17427. Date of Electronic Publication: 2023 Oct 13.
DOI: 10.1038/s41598-023-43768-6
Abstrakt: Patients suffering from painful spinal bone metastases (PSBMs) often undergo palliative radiation therapy (RT), with an efficacy of approximately two thirds of patients. In this exploratory investigation, we assessed the effectiveness of machine learning (ML) models trained on radiomics, semantic and clinical features to estimate complete pain response. Gross tumour volumes (GTV) and clinical target volumes (CTV) of 261 PSBMs were segmented on planning computed tomography (CT) scans. Radiomics, semantic and clinical features were collected for all patients. Random forest (RFC) and support vector machine (SVM) classifiers were compared using repeated nested cross-validation. The best radiomics classifier was trained on CTV with an area under the receiver-operator curve (AUROC) of 0.62 ± 0.01 (RFC; 95% confidence interval). The semantic model achieved a comparable AUROC of 0.63 ± 0.01 (RFC), significantly below the clinical model (SVM, AUROC: 0.80 ± 0.01); and slightly lower than the spinal instability neoplastic score (SINS; LR, AUROC: 0.65 ± 0.01). A combined model did not improve performance (AUROC: 0,74 ± 0,01). We could demonstrate that radiomics and semantic analyses of planning CTs allowed for limited prediction of therapy response to palliative RT. ML predictions based on established clinical parameters achieved the best results.
(© 2023. Springer Nature Limited.)
Databáze: MEDLINE
Externí odkaz:
Nepřihlášeným uživatelům se plný text nezobrazuje