he dataset includes a total of 5,249 MRI images, divided into training and validation sets. Each image is annotated with bounding boxes in YOLO format and labeled according to one of the four classes of brain tumors.
Image Characteristics
The MRI images are captured from various angles, including sagittal, axial, and coronal views. This variety ensures a comprehensive coverage of brain anatomy, enhancing the robustness of models trained with this dataset.
Annotations
A dedicated team manually annotated the bounding boxes using the LabelImg tool. This rigorous annotation process ensures high accuracy and reliability, making the dataset a valuable resource for brain tumor detection research.