<p>This study presents a comparative analysis of two distinct machine learning models, convolutional neural networks (CNN) and Random Forest (RF), in the classification of brain tumors using multi-modal MRI image and feature datasets. The objective is to address the limitations of single-modality approaches by integrating both image-based deep learning and feature-based ensemble learning within a unified framework. The initial method leverages a CNN model structured to excel at image-based recognition, with a commendable 99.80% classification accuracy on MRI scans. It applies a high-resolution layer-by-layer fine-tuning process to acquire knowledge about the rich dimensions and character of brain MRI images to prove its capability to extract and learn complex patterns. The second approach applies a RF classifier on a well pre-processed set of morphological and textural features extracted from MRI scans with 98.4% accuracy. The two models were cross-validated and statistically compared for significance to ensure reliability of the results. The model has an advantage in feature-based data with the emphasis being its speed and robustness in classifying non-image data using ensemble learning methods. The combined analysis depicts the ability of leveraging multi-modal datasets in enhancing diagnostic accuracy and reliability in medical imaging. This paper introduces a novel integrated multi-modal framework that integrates both image-based deep learning and feature-based ensemble learning for straightforward comparative and complementary evaluation on the same database. Thorough statistical analysis like cross-validation and significance testing ensures the validity of results. Contrary to previous work where one modality or approach was discussed, the current paper bridges an essential gap through the presentation of an integrated assessment of how multi-modal data fusion can make brain tumor diagnosis more reliable, interpretable, and clinically useful. The research concludes that the combined multi-modal approach not only benefits through performance but also has practical advantages for real-world clinical use.</p>