A systematic review of the hybrid machine learning models for brain tumour segmentation and detection in medical images.

Early and accurate detection of brain tumours using Magnetic Resonance Imaging (MRI) is critical for effective treatment and improved patient outcomes. This systematic review investigates the application of hybrid machine learning (ML) and deep learning (DL) models in enhancing the computational efficiency and diagnostic accuracy of brain tumour analysis from MRI images. The study synthesizes recent advances in combining traditional ML models such as Support Vector Machines (SVM) with deep neural networks like VGG-19 and YOLOv10n. A PRISMA-based literature search strategy was employed across major databases, including PubMed, Scopus, and IEEE Xplore, selecting 25 relevant studies published between 2019 and 2024. The review evaluates the performance of standalone and hybrid models using metrics such as Dice Similarity Coefficient (DSC), Intersection over Union (IoU), accuracy, precision, recall, and F1-score. Findings indicate that hybrid models, particularly those combining SVM with CNN-based architectures like VGG-19, demonstrate improved classification accuracy and reduced false positives, outperforming single-model approaches. Lightweight versions such as YOLOv10n offer faster inference times suitable for real-time applications while maintaining competitive accuracy. Despite these advances, challenges remain in model generalizability, lack of large, annotated datasets, and limited adoption of Explainable AI (XAI) for interpretability. This review highlights the potential of hybrid models for brain tumour detection and offers recommendations for future research to focus on scalable, interpretable, and clinically deployable solutions.