YOLOv3 for Brain Tumor Detection

Brain tumors are fatal diseases generally caused by abnormal cells growing in the brain. Therefore, early and accurate detection of this disease can save the patient’s life. This paper proposes a novel framework for the detection of brain tumors. The framework is based on the use of the YOLOv3 technique for detection. The proposed framework employs a deep neural network model called YOLOv3 for brain tumor detection. The Yolov3 technique is employed, using the principle of encoding bounding boxes to detect and predict the regions and types of brain tumors. Experiments with detection are performed on the Kaggle dataset of brain tumors. The results of experiments demonstrate that the proposed framework has achieved an 80% confidence score and an accuracy of 93.4% for the detection of brain tumors. The results demonstrate the efficiency of the proposed framework and have a higher performance level than other recent literature studies. Furthermore, this study is going to help in the diagnosis of brain tumors by doctors.