Published Paper Details:

Advancements in medical imaging and machine learning technologies have synergistically catalyzed breakthroughs in the field of healthcare, particularly in the early diagnosis of complex diseases such as brain tumors. This research paper presents a comprehensive investigation into the development of a robust and efficient brain tumor detection system employing state-of-the-art machine learning techniques. The proposed methodology integrates a diverse set of magnetic resonance imaging (MRI) scans, to capture a holistic representation of the brain's structural and functional aspects. A curated dataset, comprising a spectrum of brain tumor cases and healthy brain images, is utilized for training and evaluating the machine learning algorithms. Our research delves into the application of deep learning architectures, such as convolutional neural networks (CNNs) built using PyTorch, to automatically extract intricate patterns and features from the medical images. We use publicly available Kaggle dataset that has 3000 samples, which are divided into two classes tumor and non tumor. This dataset is split into 2400 images for training and 600 images for validation This allows for the creation of a highly discriminative model capable of accurately distinguishing between tumor and non-tumor regions. Our experimental results indicate that our models achieve up to 95.3 classification accuracy for our employed datasets, respectively.