Deep Learning Models Predict Brain Tumor Survival Using Population Data
-
Glioblastomas are aggressive brain tumors with poor prognosis. This study develops ML and DL models using the SEER database to predict glioblastoma survival.
-
The models utilize both classification (5 survival classes) and regression approaches for enhanced prediction. A DNN model demonstrates the best performance.
-
Model interpretability is enabled through SHAP, identifying age at diagnosis as the most important predictor.
-
This pioneering study is the first to harness DL for glioblastoma survival prediction based on SEER data.
-
The models can serve as valuable decision-support tools for tailored treatment planning to improve outcomes.