Comparative Study of Deep Learning Models for Brain Metastases Autosegmentation
Wang A, Putz F, Huang Y, Maier A (2026)
Publication Type: Conference contribution
Publication year: 2026
Journal
Publisher: Springer Science and Business Media Deutschland GmbH
Pages Range: 270-276
Conference Proceedings Title: Informatik aktuell
Event location: Lübeck
ISBN: 9783658510992
DOI: 10.1007/978-3-658-51100-5_55
Abstract
Accurate detection and segmentation of brain metastases (BM) are essential for stereotactic radiosurgery (SRS) planning. This study first systematically compares six loss functions across three representative 3D deep learning models. Results show Dice is a robust baseline, CE improves precision, and Focal has limited effectiveness. While JVSS substantially enhances small-lesion sensitivity, it reduces precision; combined losses achieve the most balanced and robust performance. Furthermore, we propose a novel inference strategy: local overlap fusion for subvolume merging (LOF_SM). By applying axis-wise local weighting in overlapping regions, LOF_SM significantly improves computational efficiency, reducing inference time by approximately 30%, while simultaneously maintaining sensitivity and improving precision over conventional sliding window (SW) methods. These findings underscore the importance of jointly optimizing loss functions and inference strategies to achieve a superior balance between detection performance and computational efficiency, thereby facilitating the clinical application of BM segmentation.
Authors with CRIS profile
Florian Putz
Department of Radiation Oncology
Andreas Maier
Lehrstuhl für Informatik 5 (Mustererkennung)
Involved external institutions
China (CN)How to cite
APA:
Wang, A., Putz, F., Huang, Y., & Maier, A. (2026). Comparative Study of Deep Learning Models for Brain Metastases Autosegmentation. In Heinz Handels, Katharina Breininger, Thomas Deserno, Andreas Maier, Klaus Maier-Hein, Christoph Palm, Thomas Tolxdorff (Eds.), Informatik aktuell (pp. 270-276). Lübeck, DE: Springer Science and Business Media Deutschland GmbH.
MLA:
Wang, Anqi, et al. "Comparative Study of Deep Learning Models for Brain Metastases Autosegmentation." Proceedings of the Bildverarbeitung für die Medizin Workshop, BVM 2026, Lübeck Ed. Heinz Handels, Katharina Breininger, Thomas Deserno, Andreas Maier, Klaus Maier-Hein, Christoph Palm, Thomas Tolxdorff, Springer Science and Business Media Deutschland GmbH, 2026. 270-276.
BibTeX: Download