Abstract—Brain Tumors (BT) is one of the major global health concerns with which diagnosis and treatment challenges are associated. Current techniques for BT identification and classification are time consuming and subjective due to manual expert interpretation of Magnetic Resonance Imaging (MRI) scans. The present research work focuses on Optimized Multi-Model Framework that incorporates DeepLoc, Machine Learning Architectures, and Ensemble Learning Technique for Enhanced Brain Tumor Classification. The framework uses DeepLoc algorithm for MRI images and exploits ML techniques including Support Vector Machine (SVM), k-Nearest Neighbor (kNN), and Naive Bayes (NB). By linking these models through ensemble techniques, we aim to attain higher classification performance. The effectiveness of approach is identified using key evaluation metrics including Training Time (TT:1.3 s), Testing Time (TET:1.9 s), Area Under Curve (AUC:88%), Classification Accuracy (CA:71%), F1-Score (F1:70%), and Recall (71%). Experimental results when compared with the previous approaches have shown a substantial increase in the values of evaluation parameters. Comprehensive validation is done using Confusion Matrix (CM), Calibration Plots (CP) and Receiver Operating Characteristic (ROC) Analysis. Ultimately, this research work intents to provide an efficient and reliable tool for early brain tumor classification, enabling timely treatment interventions for patients.
 
Keywords—Deep Learning (DL), Machine Learning (ML), Ensemble Learning (EL), brain tumor, DeepLoc


Copyright © 2025 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).