Classification and Detection of Brain Tumors by Aquila Optimizer Hybrid Deep Learning Based Latent Features with Extreme Learner

Abstract

Brain cancer is a potentially fatal illness that affects the brain. To preserve lives, early tumour detection is now crucial. Imaging in medicine is one method for diagnosing brain tumours. To help find brain tumours, researchers are turning to deep learning. Error in individual early diagnosis of the condition has been demonstrated to be reduced using deep learning algorithms. In the case of brain tumours, even a slight misdiagnosis might have serious consequences. When it comes to processing medical images, spotting brain tumours is still a difficult task. It’s difficult to spot the tumour on a brain scan. The precision of the image is impacted by many disturbances and a delay. We used deep learning methods to describe brain disorders in our paper. Brain disease detection utilizing deep learning methods is related to the study of new information. Proposed TL-based DenseNet121 model achieved accuracy, sensitivity, specificity, F1Score, precision, and IoU of 98.38, 97.33, 99.1, 98.23, 98.62, and 96.62 respectively. The results obtained on the brain tumor data set demonstrate that the proposed method outperforms others in terms of F1-score, Precision, Sensitivity, Accuracy, Specificity, and IoU.