DEEP LEARNING–BASED AUTOMATED DETECTION OF BRAIN METASTASES USING MULTICENTER MRI DATA

Brain metastases are the most common malignant brain tumors and significantly affect patient survival. Magnetic Resonance Imaging (MRI) is the primary modality for brain metastasis detection; however, manual interpretation is time-consuming and prone to diagnostic variability. This study presents a comparative evaluation of a deep learning–based MRI brain metastasis detection framework using clinical datasets from Samarkand and Germany. A 3D convolutional neural network (CNN) was trained and tested on contrast-enhanced MRI scans from both regions. Accuracy, specificity, sensitivity, F1-score, and ROC-AUC were used to evaluate performance. Due to more data diversity and established imaging techniques in Germany, deep learning produces somewhat better generalization, but the results show that deep learning achieves high diagnostic performance in both locations. The study confirms the feasibility of deploying AI-based diagnostic systems in developing and advanced healthcare environments.