Expediting Diagnostic Testing of Cancer Diagnosis Using AI-Powered Insights

Brain tumor classification is a critical task in neuro-oncology, where timely and accurate diagnosis can significantly impact treatment outcomes.This study proposes a deep learning-based approach using EfficientNetB0 to classify brain tumors from MRI images into four categories: glioma, meningioma, pituitary, and no tumor.The dataset comprises 7043 images, split into training and testing sets.The model was trained using transfer learning, followed by fine-tuning with class weighting and data augmentation.Evaluation metrics included accuracy, precision, recall, F1-score, and AUC.The model achieved a test accuracy of 82.8%, macro AUC of 0.9674, and demonstrated strong generalization across tumor types.These results highlight the potential of EfficientNetB0 as a lightweight and effective tool for assisting radiologists in clinical decisionmaking.