Evaluation of deep learning‐based methods for automatic detection and segmentation of brain metastases in T1‐contrast MRI for stereotactic radiosurgery
Journal of Applied Clinical Medical Physics · 2026
DOI: 10.1002/acm2.70459 ↗Kurz gefasst
Evaluationsstudie in der Zeitschrift Journal of Applied Clinical Medical Physics, 2026. Verfasst von Zhifeng Xu und 10 weiteren Autor:innen. Beteiligte Einrichtung: National Clinical Research Center for Cancer, Tianjin's Clinical Research Center for Cancer, Tianjin Medical University Cancer Institute & Hospital, …. Hauptthemen laut MEDLINE: Radiosurgery, Brain Neoplasms, Deep Learning, Magnetic Resonance Imaging. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 38 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Tianjin Municipal Education Commission Research Development Program, China Postdoctoral Science Foundation, Natural Science Foundation of China.
Abstract (Originalsprache)
Background
Brain metastases (BMs) are manually contoured during stereotactic radiosurgery (SRS) treatment planning, which is both time‐consuming and potentially inconsistent. To address these challenges, researchers have been actively developing deep learning‐based approaches for the detection and segmentation of BMs. However, a comprehensive comparative analysis of deep learning models across different frameworks remains largely absent in the current literature. This study aimed to evaluate and compare deep learning models based on different frameworks for the detection and segmentation of BMs in T1‐contrast MRI. Materials and Methods Eight deep learning models, based on CNN, Transformer, or Mamba architectures, were trained and validated for the task of detecting and segmenting brain metastatic lesions in T1‐contrast MRI. A total of 934 patients were included, with 667 cases from publicly available datasets and 267 cases from our institution, designated for training and testing, respectively. Data were retrospectively collected and organized at our institution, and GTV defined as the total BM tumor volume delineated by the physician at the time of stereotactic radiosurgery (SRS). Additionally, labels in the publicly available dataset were modified under clinician guidance to create a BM GTV that met clinical criteria to improve ground‐truth accuracy. A BM was considered detected if the ground‐truth contour overlapped with a predicted structure. Sensitivity at both the patient‐level (proportion of patients with at least one lesion detected) and lesion‐level (proportion of ground‐truth lesions detected) were used to evaluate BM detection. Segmentation performance was assessed using several metrics: dice similarity coefficient (DSC), positive predictive value (PPV), surface DSC (sDSC), and Hausdorff distance 95% (HD95). The performance across different BM diameters was also evaluated.
Results
Among the eight deep learning models, the U‐Mamba (Bot) achieved a lesion‐level sensitivity of 0.796 (95% CI: 0.779–0.812) for all sizes of BM, which was significantly higher than that of the other models, with a false positive rate of 2.46 ± 4.96 per patient. Further stratification by metastasis diameter, the sensitivity was 0.505 for BMs < 3 mm, 0.797 for BMs between 3 and 6 mm, and 0.885 for BMs between 6 and 9 mm. Moreover, U‐Mamba (Enc) demonstrated significantly higher lesion‐level segmentation performance, with DSC value of 0.632 ± 0.224. In terms of tumor boundary segmentation, nnU‐Netv2 achieved the best performance, with Surface DSC and HD95 values of 0.877 ± 0.149 and 1.770 ± 1.458 mm.
Conclusion
The nnU‐Netv2 allows precise segmentation of lesion areas in T1‐contrast MRI, while U‐Mamba provide effective detection of brain metastasis, potentially aiding in treatment planning for SRS.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- National Clinical Research Center for Cancer, Tianjin's Clinical Research Center for Cancer, Tianjin Medical University Cancer Institute & Hospital, Tianjin, China.
- School of Biomedical Engineering and Technology, Tianjin Medical University, Tianjin, China.
ORCID-Profile: Zhifeng Xu, Yi Xue, Yang Dong
Förderung
- Tianjin Municipal Education Commission Research Development Program
- China Postdoctoral Science Foundation
- Natural Science Foundation of China
- Tianjin Key Medical Discipline Construction Project
- National Natural Science Foundation of China
Zitieren
APA
Xu, Z., Yang, Y., Wang, G., Xue, Y., Zhang, X., Dong, Y., Xu, L., Wang, Q., Wang, W., Yuan, Z., & Huang, S. (2026). Evaluation of deep learning‐based methods for automatic detection and segmentation of brain metastases in T1‐contrast MRI for stereotactic radiosurgery. Journal of Applied Clinical Medical Physics. https://doi.org/10.1002/acm2.70459
Vancouver
Xu Z, Yang Y, Wang G, Xue Y, Zhang X, Dong Y, et al. Evaluation of deep learning‐based methods for automatic detection and segmentation of brain metastases in T1‐contrast MRI for stereotactic radiosurgery. Journal of Applied Clinical Medical Physics. 2026. doi:10.1002/acm2.70459
BibTeX
@article{Xu2026,
author = {Xu, Zhifeng and Yang, Yuqi and Wang, Guanjie and Xue, Yi and Zhang, Xinyang and Dong, Yang and Xu, Liming and Wang, Qi and Wang, Wei and Yuan, Zhiyong and Huang, Sheng},
title = {Evaluation of deep learning‐based methods for automatic detection and segmentation of brain metastases in T1‐contrast MRI for stereotactic radiosurgery},
journal = {Journal of Applied Clinical Medical Physics},
year = {2026},
doi = {10.1002/acm2.70459}
}
EUVIMED