Brain cancer is one of the deadliest diseases affecting the brain. Medical imaging techniques are commonly employed for the identification of brain tumours, and Deep Learning is rapidly emerging as a powerful tool in this detection process. This study evaluates the efficiency of Deep Learning Convolutional Neural Networks (CNNs)—including ResNet50, EfficientNetB0, VGG16, and InceptionV3—against traditional Machine Learning models such as Logistic Regression, Support Vector Machine (SVM), and Random Forest for detecting brain tumours from Magnetic Resonance Imaging (MRI) scans. Automated detection techniques offer significant improvements in patient outcomes, particularly given the increasing incidence of brain tumours and the urgent need for swift, accurate diagnoses to save lives. To address this challenge, the study utilised advanced CNN architectures to analyse the “Br35H: Brain Tumour Detection 2020” dataset of brain MRI images with annotations. The performance of the models was assessed based on accuracy, precision, recall (sensitivity), specificity, F1-score, and ROC– AUC metrics. The results demonstrated that Deep Learning models achieved significantly higher accuracy compared to traditional Machine Learning methods, with ResNet50 yielding the best results: 99.33% accuracy, 99.11% precision, and 99.55% recall.