Answers are grounded in this paper and cite their sources.
Brain tumor classification and detection from Magnetic Resonance Imaging(MRI) images is critical for early diagnosis. State-of-the-art deep learning (DL) methods, such as the You Only Look Once (YOLO) family of detectors, have recently gained widespread success in analysis of medical images. In this paper, based on the state-of-the-art YOLOv11 and YOLOv8 architectures, The study has propose a brain tumor detection framework and improve the performance of brain tumor detection by replacing the conventional Cross Stage Partial(CSP)-based backbone with a Pyramid Vision Transformer (PVT). The baseline YOLOv11 model is trained with a balanced and augmented Brain Tumor MRI Dataset, achieving improved Precision, Recall, F1-score, Mean Average Precision(mAP@0.5), and mAP@0.5:0.95 compared with baseline YOLOv8 and YOLOv11 models. To overcome the limitations of convolution-based feature extraction, particularly the lack of global contextual information and difficulty in detecting small or low-contrast tumors, this study proposes a hybrid YOLOv11-PVT model. The PVT backbone generates hierarchical multi-scale features and global attention representations, thereby providing richer contextual information. Results show that the proposed YOLOv11-PVT framework demonstrated improved sensitivity and stronger IoU-based localization consistency compared with the baseline YOLOv11 model. These results demonstrate that transformer-enhanced backbones can improve the robustness and localization capability of tumor detection systems.
No takes yet. Share an insight, caveat, or question.
Srinivasu et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: