Abstract Objectives: The primary objective of the research is to assess the effectiveness of deep learning model for automatic brain tumor detection from MRI images. The objective is to assess its ability to classify cases as tumor or non-tumor with great precision and dependability with accuracy. Method: The Pre trained Convolutional Neural Network (CNN) framework was employed on MRI scans for classification technique. The dataset employed in this study comprised of 1 MRI image, which underwent preprocessing (resizing and normalization) prior to prediction. The result is a probability measure showing the likelihood of the image displaying a tumor or non-tumor. Findings: Based on the evaluation, this model accurately categorized 1 case as tumor and 0 cases as non-tumor. The model achieved an average tumor probability of 99.74%, highlighting its confidence in prediction. This model reaches an accuracy of 80%, with a precision of 85%, a recall of 75%, and an F1 score of 79.5%. Novelty: This research demonstrates the successful application of a CNN-based brain tumor detection system that employs real MRI data, highlighting quantifiable prediction accuracy. The breakthrough lies in its ability to provide precise forecasts with strong confidence, even with limited input data, facilitating incorporation into clinical decision supports systems. Keywords: Brain Tumor Detection, MRI Classification, Deep Learning, Convolutional Neural Network (CNN), Evaluation of Medical Images, Tumor Prediction, Computer-Aided Diagnosis, Artificial Intelligence in Healthcare, Image Assessment, Forecasting Analytics
Vishnuvardhan et al. (Fri,) studied this question.