Analysis of Deep Learning Models for Brain Tumour Detection

The detection and identification of brain tumors is vital for facilitating early diagnosis and enabling timely treatment of such conditions. This paper investigates the utilization of deep learning architectures, including Convolutional Neural Networks (CNN) [9], VGG [10], U-Net [7], Transfer Learning [11] and RU-Net [12] for tumors cerebral detected MRI. imagine The models are trained using the Kaggle dataset brain images for brain tumor detection dataset, which contains images labeled “yes” (tumor) and “no” (no tumor). The efficacy of each model is assessed according to parameters such as accuracy, loss and generalization. The outcomes indicate that the CNN achieved a test accuracy of 88. 23%, while Transfer Learning produced an accuracy of 82.35% on unseen data. The performance of the models is compared and suggestions for improvement are given, including adjustment and addition of data.