Convolutional Neural Network based Brain Tumor Detection
Shraddha S. More, Mansi Ashok Mange, Mitheel Sandip Sankhe, Shwethali Santosh Sahu
Abstract
One of the most important and demanding tasks in the field of medical image processing is Brain Tumor Detection, as inaccurate prediction and diagnosis can result from human-assisted manual classification. One of the functions of Artificial Intelligence is Deep Learning which mimics the work of the person's brain. It is used to detect artifacts in the processing of data, recognize the voice, translate languages, and make decisions. Without human administration, it may understand, demonstrating from data that is both unorganized and unlabelled. A Convolutional Neural Network is a form of deep neural network which is used most commonly in optical representation analysis in deep learning. The current situation provides systems that detect brain tumors, but only use small datasets and image processing techniques. The proposed system majorly consists of 3 parts namely: Augmentation, Image pre-processing and applying Con volutional Neural Network (CNN). Our approach is to propose a system in which we will use a large dataset and deep learning algorithm. Results demonstrate thatthe CNN has 87.42 %training accuracy with low difficulty, which sets it apart from all other state-of-the-art approaches.