الكشف عن وتشخيص أورام المخ باستخدام الشبكات العصبية التلافيفية

Abstract

Brain tumors are one of the major health problems related to abnormalities in the human brain. An activist diagnosis of brain tumors is critical to improve patient outcomes and lives. Early detection of tumors is crucial for treatment. Magnetic Resonance Imaging (MRI) is one of the most commonly used diagnostic methods for brain tumors in the clinical area. Manual detection of brain tumors is becoming increasingly time-consuming and costly. Therefore, an automated Computer Aided Diagnosis (CAD) system is needed to help doctors and radiologists detect these deadly tumors in time, thereby saving precious lives. Convolutional Neural Networks (CNNs) are widely used in various CAD systems. CNNs play an important role in healthcare as image processing techniques for segmentation, recognition, and classification of MRI images and classification and detection of brain tumors. This paper applies the Deep Learning (DL) architectures for brain tumor detection and classification. Implement a deep learning (CNN) based computational approach that includes image pre-processing to extract regions of interest in the image itself to identify and detect tumors in the brain. The highest accuracy rate in the experiment reached 96%. Evaluation metrics used include sensitivity, precision, loss, and F1 score. The results obtained will aid in the diagnosis and detection of brain tumors.