Abstract:

Accurate classification of brain tumors plays a vital role in clinical diagnosis and therapy. To aid in this process, we present a deep learning approach for brain tumor classification. By leveraging deep learning, radiologists can efficiently analyze the vast amount of brain MRI images, leading to faster and more accurate diagnoses. However, training deep learning models requires large centralized datasets, which can pose challenges due to privacy regulations surrounding medical data. In this study, we address this issue by developing a model that utilizes a Generative Adversarial Network (GAN) to generate synthetic brain tumor MRI images. Additionally, we propose a hybrid CNN-LSTM network to accurately identify brain tumors in MRI scans. The performance of the hybrid network achieves an impressive classification accuracy of 99.1%.

Description:

Automated Brain Tumor detection, GAN Network, deep neural network