Towards Radiomics-based automated disease progression assessment for glioblastoma patients

Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries. 9th International Workshop, BrainLes 2023, and 3rd International Workshop, SWITCH 2023, Held in Conjunction with MICCAI 2023, Vancouver, BC, Canada, October 8 and 12, 2023, Revised Selected Papers

Themenheft

DOI der Originalpublikation

https://doi.org/10.1007/978-3-031-76160-7_4

URI

https://irf.fhnw.ch/handle/11654/54728

Link

Reihe / Serie

Lecture Notes in Computer Science

Reihennummer

14668

Jahrgang / Band

Ausgabe / Nummer

Seiten / Dauer

36-47

Patentnummer

Verlag / Herausgebende Institution

Springer

Verlagsort / Veranstaltungsort

Vancouver

Auflage

Version

Programmiersprache

Abtretungsempfänger:in

Praxispartner:in/Auftraggeber:in

Zusammenfassung

Glioblastoma is a highly infiltrative brain tumor with fast progression and poor prognosis for patients. Due to the rapid growth, close treatment response monitoring is key. In this study, we benchmark different machine learning approaches for automated progression classiu0002fication with various radiomic feature sets extracted from longitudinal magnetic resonance imaging and classifiers. Our experiments show difu0002ferences in robustness and performance and offer insights into common failure modes. The best ROC-AUC was achieved with a random forest classifier without feature selection (0.748), and the best F1 score was at 0.792 for an XGBoost classifier where features of the current time point and the change from the reference time point were provided. Anau0002lyzing misclassifications shows different behavior for statistical machine learning classifiers and Residual Neural Networks.

Schlagwörter

Fachgebiet (DDC)

330 - Wirtschaft

Projekt

Veranstaltung

The 26th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2023)

Startdatum der Ausstellung

Enddatum der Ausstellung

Startdatum der Konferenz

08.10.2023

Enddatum der Konferenz

12.10.2023

Datum der letzten Prüfung

ISBN

978-3-031-76159-1
978-3-031-76160-7

ISSN

Sprache

Englisch

Während FHNW Zugehörigkeit erstellt

Nein

Zukunftsfelder FHNW

Publikationsstatus

Veröffentlicht

Begutachtung

Peer-Review der ganzen Publikation

Open Access-Status

Closed

Lizenz

Zitation

Suter, Y., Schuhmacher, F., Ermis, E., Knecht, U., Schucht, P., Wiest, R., & Reyes, M. (2024). Towards Radiomics-based automated disease progression assessment for glioblastoma patients. In U. Baid, R. Dorent, S. Malec, M. Pytlarz, R. Su, N. Wijethilake, S. Bakas, & A. Crimi (Eds.), Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries. 9th International Workshop, BrainLes 2023, and 3rd International Workshop, SWITCH 2023, Held in Conjunction with MICCAI 2023, Vancouver, BC, Canada, October 8 and 12, 2023, Revised Selected Papers (pp. 36–47). Springer. https://doi.org/10.1007/978-3-031-76160-7_4

Komplettanzeige