XGBoost-Powered Hybrid Ensemble Approach for Precise Brain Tumor Diagnosis, Combining Convolutional Neural Networks, ResNet50 and K-Means Clustering Techniques

Brain tumor categorization is an important issue in medical imaging because it helps clinicians diagnose and plan effective treatments. In order to improve brain tumor detection systems' accuracy and resilience, this study presents a novel hybrid ensemble framework. Three essential elements are integrated into the suggested methodology. First, an array of machine learning models including Convolutional Neural Network (CNN), Support Vector Machine (SVM), and Extreme Gradient Boosting (XGBoost) is used for dependable classification. Second, Residual Network with 50 layers (ResN et50) for extracting deep, discriminative features. Third, K-Means clustering for efficient tumor segmentation. By separating tumor regions, K-Means clustering enhances segmentation and facilitates improved feature representation. The complex patterns and textures of the tumor are captured by the hierarchical features extracted by the pre-trained deep learning model ResNet50. Lastly, the ensemble model ensures consistency over a range of data by utilizing the capabilities of several classifiers to lower the risks of bias and overfitting. The architecture of the proposed system tackles a number of issues, such as class imbalance, noisy data, and the wide range of tumor presentations in medical imaging. The suggested method provides a scalable, automated workflow with little manual intervention by fusing unsupervised clustering with sophisticated feature extraction and ensemble learning. The framework's ability to differentiate between tumor and healthy photos is demonstrated by the experimental findings on a carefully selected dataset, offering important information for clinical decision-making. This work supports continued attempts to create reliable, AI-powered medical diagnostic tools. Keywords: Brain tumor classification, CNN, hybrid ensemble framework, K-Means clustering, ResNet50, support vector machine, XGBoost.