Predicting Overall Survival in Brain Metastasis Patients Using MRI Morphological Features: A Machine Learning and Survival Analysis Approach

Background/Objectives: Our study aims to identify potential new MRI features of brain metastases (BMs) that could be further used in overall survival (OS) assessment. Methods: A total of 109 patients with BMs were included. Kaplan-Meier analysis, the log-rank test, and Cox regression were implemented in the survival analysis. Signifi-cant features were subsequently incorporated into four distinct machine learning (ML) algorithms to predict six-month survival. Results: Survival analysis revealed that multiple brain lesions, synchronous presentation, and restricted diffusion were associ-ated with a poor prognostic value (HR > 1; p = 0.01, p = 0.02, and p < 0.005, respective-ly). Other features demonstrated a protective effect OS: the absence of extracranial le-sions (HR < 1, p = 0.04), and the presence of peripheral edema and solid enhancement (HR < 1, p < 0.0005). All treatment protocols—including surgery, gamma knife radio-surgery, whole brain radiation therapy, and chemotherapy - showed significantly bet-ter survival rates compared to best supportive care (HR < 1, p < 0.005). The Neural Networks was the top-performing ML model, achieving an AUC of 0.93. According to the Shapley Additive Explanations analysis, solid enhancement and radiotherapy had a positive impact on OS, whereas a higher number of lesions, larger volumes, and re-stricted diffusion were associated with negative outcomes. Conclusions: Our results confirm that including morphological MRI features of BMs in the prediction of OS sig-nificantly contributes to the enhancement of ML algorithms prediction and discrimi-natory capacity.