A Deep Learning Approach for Brain Tumor Segmentation in MRI Images Using YOLOv7

Ramya Nimmagadda^*Kalpana Devi P

• Vel Tech Rangarajan Dr.Sagunthala R&D Institute of Science and Technology, Chennai, India

The medical imaging field has seen tremendous growth as a result of the latest advancements in digital imaging and artificial intelligence (AI). These advancements have improved the tumor classification accuracy, time, cost efficiency and many more. Radiologists utilize a MRI scan due to its exceptional capacity to identify even the smallest alterations in brain activity. This research uses YOLOv7, a Deep Learning (DL) model, for the classification and detection of brain tumors and to conduct a detailed analysis of the frequently used structures for tumor identification. There are four distinct classes in our brain tumor dataset: pituitary tumors, gliomas, meningiomas, and no tumors. This preprocessed sample was used to assess the performance of deep learning models on identifying and classifying brain tumors. Throughout the preprocessing stage, aspect ratio normalization and resizing algorithms are applied to the model to ensure precise tumor segmentation. YOLOv7 performs admirably with a recall score of 0.813 and a box detection accuracy of 0.837. Remarkably, the mAP value for the 0.5 IoU threshold is 0.879. During box identification within the extended IoU spectrum of 0.5 for a to 0.95, the mAP value was 0.442.