Brain Tumor Detection Using CNN: A Deep Learning Approach for MRI Classification

A major focus of medical diagnostics is brain tumor detection; early and correct diagnosis greatly enhances patient outcomes here. This paper provides a Convolutional Neural Network (CNN)-based method to classify brain MRI images as either positive (tumor found) or negative (not tumor detected). The suggested model was developed by means of methodical training, validation, and assessment applied on a 3,264 MRI open-source dataset from the Kaggle portal. Resizing, normalizing, and augmenting data helped to preprocess the dataset so improving model generalizing capability. After an 80–20 data split and training and validation, the model obtained 82% on the test set. With a precision of 0.85, recall of 0.84, and an F1-score of 0.85 for positive cases the classification report shows strong performance. Comparatively, for negative cases, the precision, recall, and F1-score were 0.78, 0.79, and 0.79, respectively, thereby guaranteeing consistent classification over both classes. Emphasizing its potential as a decision-support tool in clinical environments, the suggested CNN model efficiently and highly accurately automates brain tumor identification. Future research aims to maximize model performance for imbalanced datasets and use explainability methods to build medical practitioner trust.