Hybrid Deep Learning for Survival Prediction in Brain Metastases Using Multimodal MRI and Clinical Data
Diagnostics · 14.05.2025 · 15 (10) · 1242
DOI: 10.3390/diagnostics15101242 ↗Kurz gefasst
Fachartikel in der Zeitschrift Diagnostics, online erschienen am 14.05.2025. Verfasst von Cristian Constantin Volovăț und 10 weiteren Autor:innen. Beteiligte Einrichtung: Radiology Department, "Grigore T. Popa" University of Medicine and Pharmacy Iași, 700115 Iași, Romania. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 43 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Background
Survival prediction in patients with brain metastases remains a major clinical challenge, where timely and individualized prognostic estimates are critical for guiding treatment strategies and patient counseling.
Methods
We propose a novel hybrid deep learning framework that integrates volumetric MRI-derived imaging biomarkers with structured clinical and demographic data to predict overall survival time. Our dataset includes 148 patients from three institutions, featuring expert-annotated segmentations of enhancing tumors, necrosis, and peritumoral edema. Two convolutional neural network backbones—ResNet-50 and EfficientNet-B0—were fused with fully connected layers processing tabular data. Models were trained using mean squared error loss and evaluated through stratified cross-validation and an independent held-out test set.
Results
The hybrid model based on EfficientNet-B0 achieved state-of-the-art performance, attaining an R2 score of 0.970 and a mean absolute error of 3.05 days on the test set. Permutation feature importance highlighted edema-to-tumor ratio and enhancing tumor volume as the most informative predictors. Grad-CAM visualizations confirmed the model’s attention to anatomically and clinically relevant regions. Performance consistency across validation folds confirmed the framework’s robustness and generalizability.
Conclusions
This study demonstrates that multimodal deep learning can deliver accurate, explainable, and clinically actionable survival predictions in brain metastases. The proposed framework offers a promising foundation for integration into real-world oncology workflows to support personalized prognosis and informed therapeutic decision-making.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Radiology Department, "Grigore T. Popa" University of Medicine and Pharmacy Iași, 700115 Iași, Romania.
- "Prof. Dr. Nicolae Oblu" Clinical Emergency Hospital Iași, 700309 Iași, Romania.
- National Institute of Research and Development for Technical Physics, IFT Iași, 700050 Iași, Romania.
- Medical Oncology-Radiotherapy Department, "Grigore T. Popa" University of Medicine and Pharmacy Iași, 700115 Iași, Romania.
- Physics Department, "Gheorghe Asachi" Technical University Iași, 700050 Iași, Romania.
- Faculty of Pharmacy, "Grigore T. Popa" University of Medicine and Pharmacy Iași, 700115 Iași, Romania.
ORCID-Profile: Călin Gheorghe Buzea, Diana-Ioana Boboc, Lăcrămioara Ochiuz, Dragoș Ioan Rusu
Zitieren
APA
Volovăț, C. C., Buzea, C. G., Boboc, D. I., Ostafe, M. R., Agop, M., Ochiuz, L., Burlea, Ș. L., Rusu, D. I., Bujor, L., Iancu, D. T., & Volovăț, S. R. (2025). Hybrid Deep Learning for Survival Prediction in Brain Metastases Using Multimodal MRI and Clinical Data. Diagnostics, 15(10), 1242. https://doi.org/10.3390/diagnostics15101242
Vancouver
Volovăț CC, Buzea CG, Boboc DI, Ostafe MR, Agop M, Ochiuz L, et al. Hybrid Deep Learning for Survival Prediction in Brain Metastases Using Multimodal MRI and Clinical Data. Diagnostics. 2025;15(10):1242. doi:10.3390/diagnostics15101242
BibTeX
@article{Volov2025,
author = {Volovăț, Cristian Constantin and Buzea, Călin Gheorghe and Boboc, Diana-Ioana and Ostafe, Mădălina-Raluca and Agop, Maricel and Ochiuz, Lăcrămioara and Burlea, Ștefan Lucian and Rusu, Dragoș Ioan and Bujor, Laurențiu and Iancu, Dragoș Teodor and Volovăț, Simona Ruxandra},
title = {Hybrid Deep Learning for Survival Prediction in Brain Metastases Using Multimodal MRI and Clinical Data},
journal = {Diagnostics},
year = {2025},
volume = {15},
number = {10},
pages = {1242},
doi = {10.3390/diagnostics15101242}
}
EUVIMED