Machine Learning Techniques For Automated And Early Detection Of Brain Tumor

Abstract

A tumour, as the name implies, is a tumorous growth of tissue anywhere in the body. There are various types of tumours, each with its own set of characteristics and treatment plan. The goal of this study is to create a reliable algorithm for detecting tumours in brain MRI images. Image segmentation is critical for detecting brain tumours. One of the most difficult, but crucial, processes is detecting a brain tumour. As a result, accurate segmentation of Magnetic Resonance Imaging (MRI) images is critical for subsequent diagnosis. The ongoing research into automatic detection of brain structures is motivated by a desire to learn more about the connections between the anatomy of brain tissues and various mental and physical disorders in humans. These days, medical professionals are particularly interested in computer-aided technologies that can identify and characterise certain organs or medical characteristics. Using image processing and machine learning, this study proposes a strategy for the early and accurate detection of brain tumours. The SVM, ANN, and ID3 algorithms are all utilised in some capacity within the context of this framework's procedures for extracting features and segmenting images. Metrics such as accuracy, specificity, and sensitivity are utilised in the evaluation process so that we can determine how well an algorithm performs.