A Deep Learning Framework For Rapid And Automated Brain Tumor Classification: The CNN-Based Diagnostic Platform

Accurate and timely diagnosis of brain tumors—specifically Glioma, Meningioma, and Pituitary Tumor—is a critical challenge in clinical neurology. Traditional Magnetic Resonance Imaging (MRI) analysis is highly dependent on radiologist expertise, leading to potential variability and diagnostic delays. This research introduces a novel, end-to-end framework for a Deep Learning and CNN-based Brain Tumor Detection Platform. The proposed system, termed the Automated Neuro-Diagnostic Assistant (ANDA), is built on a custom Convolutional Neural Network (CNN) architecture trained on pre-processed MRI datasets. The core innovation lies in its deployment as an interactive, real-time Flask web platform that integrates the model with features like a Confidence Score Visualizer and a Digital Report Generator. Preliminary validation demonstrates 96% accuracy and a high F1-Score of 94% on the test dataset , effectively establishing a paradigm shift from manual image interpretation to an active, cognitive partner in neuro-radiology.