<p>Brain tumours remain a critical health challenge, necessitating advancements in diagnostic and treatment methodologies. This research explores innovative approaches to improve accuracy and efficiency in brain tumor management, addressing limitations in current practices. Existing diagnostic methods often face challenges such as low accuracy, high false-positive rates, and time-consuming processes, which can delay effective treatment. To overcome these issues, this study proposes a novel framework called Central Nervous System using Machine Learning (CNS-ML), which integrates advanced machine learning techniques for enhanced tumor detection and classification. The proposed method utilizes Random Forest, a robust classification algorithm, to analyze medical imaging data and improve diagnostic precision. This study introduces CNS-ML, a streamlined Random Forest-based framework that integrates optimized preprocessing and radiomic feature extraction to classify glioma, meningioma, pituitary, and no-tumor classes from MRI images with high reliability. The study significantly improves tumor detection accuracy by leveraging CNS-ML, reducing false positives and processing time. The findings demonstrate that the proposed framework outperforms traditional methods, offering clinicians a reliable and efficient tool. Ensemble learning is utilized and multi-parametric data (T1/T2/FLAIR) are processed, thus this method eliminates the drawbacks that come with conventional MRI analysis.</p>