Radiomics-based prediction of local tumor control in patients with brain metastases after stereotactic radiosurgery

Abstract

Background: Stereotactic radiosurgery (SRS) is a standard treatment for brain metastases. However, prediction of local tumor control remains challenging. We developed a radiomics-based model to predict local control after SRS. Methods: We analyzed pre-treatment MRI scans of patients with brain metastases treated with SRS. Radiomic features were extracted and machine learning models were trained to predict 12-month local control. Results: The radiomics model showed good predictive performance (AUC 0.78) and outperformed clinical-only models. Conclusion: Radiomics-based prediction may support treatment decisions in patients with brain metastases.

Kristian Unger

Professor of AI assisted therapy decisions in oncology

Multi-omics integration and prediction