Paper Key : IRJ************787
Author: Kartiki Yuvraj Vedpathak,Gauri Sunil Jadhav
Date Published: 14 Apr 2025
Abstract
This research aims to evaluate and compare the performance of YOLO-based deep learning models (from YOLOv3 to YOLOv11) for the detection and classification of brain tumors in MRI images. The goal was to determine which YOLO version is most accurate and efficient in identifying and categorizing tumor types such as gliomas, meningiomas, and pituitary tumors. MRI scans were preprocessed and passed through each YOLO model, and performance was measured using precision, recall, and F1-score.The analysis showed progressive improvements across YOLO versions, with YOLOv7 and YOLOv11 demonstrating the highest accuracy. YOLOv7 was effective due to its integration of CBAM and SPPF+ layers, enabling precise localization, while YOLOv11 outperformed all others by leveraging self-supervised learning and advanced attention mechanisms. These models consistently provided reliable results in detecting small and complex tumors.The findings are significant because early and accurate brain tumor detection is critical for timely treatment and better patient outcomes. By identifying the most effective YOLO architecture for medical image analysis, this research supports the development of faster, more accurate, and automated diagnostic tools that can assist radiologists and reduce diagnostic errors in clinical practice.
DOI Requested