Brain Tumor Identification using Improved YOLOv8

2025-02-06

AI Summary

Manual delineation of tumor boundaries in brain MRI scans is time-consuming and heavily reliant on expert radiological expertise. To address this, this paper proposes an end-to-end detection framework based on an enhanced YOLOv8 architecture. Key contributions include: (1) replacing conventional non-maximum suppression (NMS) with RT-DETR to improve robustness in boundary localization; (2) integrating Ghost convolution into the backbone to reduce computational redundancy while preserving feature expressiveness; and (3) embedding a lightweight Vision Transformer module to strengthen long-range contextual modeling. The model is trained and validated on a public brain tumor MRI dataset. Quantitative evaluation yields mAP@0.5 = 0.91—outperforming the original YOLOv8 and ten state-of-the-art detectors—demonstrating both high accuracy and real-time inference capability. This advancement offers a reliable, clinically deployable solution for computer-aided diagnosis of brain tumors.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Hardware-aware ML
• Humans and AI: Brain-Sensing and Analysis

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Identifying the extent of brain tumors is a significant challenge in brain cancer treatment. The main difficulty is in the approximate detection of tumor size. Magnetic resonance imaging (MRI) has become a critical diagnostic tool. However, manually detecting the boundaries of brain tumors from MRI scans is a labor-intensive task that requires extensive expertise. Deep learning and computer-aided detection techniques have led to notable advances in machine learning for this purpose. In this paper, we propose a modified You Only Look Once (YOLOv8) model to accurately detect the tumors within the MRI images. The proposed model replaced the Non-Maximum Suppression (NMS) algorithm with a Real-Time Detection Transformer (RT- DETR) in the detection head. NMS filters out redundant or overlapping bounding boxes in the detected tumors, but they are hand-designed and pre-set. RT-DETR removes hand-designed components. The second improvement was made by replacing the normal convolution block with ghost convolution. Ghost Convolution reduces computational and memory costs while maintaining high accuracy and enabling faster inference, making it ideal for resource-constrained environments and real-time applications. The third improvement was made by introducing a vision transformer block in the backbone of YOLOv8 to extract context-aware features. We used a publicly available dataset of brain tumors in the proposed model. The proposed model performed better than the original YOLOv8 model and also performed better than other object detectors (Faster R- CNN, Mask R-CNN, YOLO, YOLOv3, YOLOv4, YOLOv5, SSD, RetinaNet, EfficientDet, and DETR). The proposed model achieved 0.91 mAP (mean Average Precision)@0.5.

Problem

• Accurate brain tumor detection in MRI images.
• Improving YOLOv8 for real-time tumor identification.
• Reducing computational costs in tumor detection models.

Innovation

• Modified YOLOv8 for MRI tumor detection
• Replaced NMS with RT-DETR in detection head
• Introduced Ghost Convolution and Vision Transformer