Advancing Brain Tumor Detection in MRI Images Using Convolutional Neural Networks and Deep Learning Techniques

Brain cancer remains a major health challenge, particularly in constraint-driven settings like India, where it is the 10th leading cause of cancer-related deaths. Early detection is vital to improving treatment outcomes, yet traditional methods relying on manual MRI interpretation can be time taking, expensive, and highly susceptible to error. This study is presenting a deep learning based approach which is using Convolutional Neural Networks (CNNs). The proposed system uses advanced data preprocessing techniques, including Z-score normalization, skull stripping, and augmentation, to optimize MRI datasets. A robust CNN model was developed and trained on a diverse dataset comprising glioma, meningioma, pituitary, and non-tumor cases. This model achieved exceptional performance, with training, validation, and test accuracies of 99.95%, 99.8%, and 99.3%, respectively, along with high accuracy, recall, and F1 points. Evaluation metrics such as ROC-AUC and confusion matrix analysis showcased the model's accuracy in differentiating tumour types and stages.