A Two-Tier Framework Based on GoogLeNet and YOLOv3 Models for Tumor Detection in MRI

Medical Image Analysis (MIA) is one of the active research areas in computer vision, where brain tumor detection most investigated domain among researchers due to its deadly nature. Brain magnetic resonance imaging (MRI) assists radiologists for better analysis about exact size and location tumor. However, existing systems may not efficiently classify human tumors with significantly higher accuracies. In addition, smart easily implementable approaches are unavailable 2D 3D medical images, which main problem detecting this paper, we investigate various deep learning models localization MRI. A novel two-tier framework proposed first tire classifies normal MRI followed by regions second tire. Furthermore, introduce a well-annotated dataset comprised images. The experimental results demonstrate effectiveness achieving 97% accuracy using GoogLeNet on classification 83% tasks after fine-tuning pre-trained you only look once (YOLO) v3 model.