SCC-YOLO: An Improved Object Detector for Assisting in Brain Tumor Diagnosis

TL;DR

This paper introduces SCC-YOLO, an enhanced object detection model for brain tumor diagnosis that integrates SCConv into YOLOv9, improving accuracy and efficiency in medical imaging tasks.

Contribution

The paper proposes SCC-YOLO, a novel architecture that combines SCConv with YOLOv9, achieving improved detection performance for brain tumors in medical images.

Findings

SCC-YOLO increased mAP50 by 0.3% on Br35H dataset.

SCC-YOLO increased mAP50 by 0.5% on custom dataset.

Achieved state-of-the-art results in brain tumor detection.

Abstract

Brain tumors can lead to neurological dysfunction, cognitive and psychological changes, increased intracranial pressure, and seizures, posing significant risks to health. The You Only Look Once (YOLO) series has shown superior accuracy in medical imaging object detection. This paper presents a novel SCC-YOLO architecture that integrates the SCConv module into YOLOv9. The SCConv module optimizes convolutional efficiency by reducing spatial and channel redundancy, enhancing image feature learning. We examine the effects of different attention mechanisms with YOLOv9 for brain tumor detection using the Br35H dataset and our custom dataset (Brain_Tumor_Dataset). Results indicate that SCC-YOLO improved mAP50 by 0.3% on the Br35H dataset and by 0.5% on our custom dataset compared to YOLOv9. SCC-YOLO achieves state-of-the-art performance in brain tumor detection.