Published Paper Details:

Brain tumors represent a significant global health challenge, affecting approximately one million individuals worldwide in 2023. This study evaluates the efficacy of the object detection algorithm for the detection and categorization of brain tumors using MRI images. The evaluation utilized a dataset of MRI images from different patients, covering three primary brain tumor types: Pituitary, Meningioma, and Glioma. The model demonstrated robust performance, with general high precision, recall, and mAP values on the validation dataset. Notably, the model effectively differentiated between the various brain tumor types and background, as evidenced by the normalized confusion matrix. Practical testing on brain MRI images confirmed the model's ability to accurately identify different tumor regions. In conclusion, the object detection algorithm presents a promising automated method for identifying brain tumors and classification in MRI images, potentially aiding clinicians in precise diagnosis and treatment planning. Further research using larger datasets is recommended to enhance the model's clinical applicability and reliability.