Automatic Brain Tumor Segmentation

Automatic brain tumor segmentation and classification play a crucial role in early detection and diagnosis of brain-related diseases, significantly improving treatment outcomes. This study focuses on the development of an automated system for brain tumor classification and segmentation using advanced deep learning models. The proposed system integrates two core components: classification and segmentation, designed to accurately identify and delineate brain tumors in medical images.

For classification, convolutional neural networks (CNN) and MobileNet architectures are employed to distinguish between four tumor categories: Glioma, Meningioma, Pituitary tumor, and No tumor. CNN is selected for its ability to learn hierarchical features from medical images, while MobileNet offers a lightweight alternative with computational efficiency, making it suitable for real-time applications on edge devices. These models are trained using a dataset of brain MRI images, with the goal of achieving high accuracy in tumor classification.

Keywords: Brain tumor segmentation, classification, CNN, MobileNet, UNet++, Glioma, Meningioma, Pituitary tumor, deep learning, medical image analysis.