Detection of Brain Tumor in Medical Images Based on Feature Extraction by HOG and Machine Learning Algorithms
Jayaraj Ramasamy, Ruchi Doshi, Kamal Kant Hiran
Abstract
Computer-aided diagnostics play a big part in figuring out what’s wrong with MRI images, which helps the radiologist. As among the most prevalent and incurable cancers, brain tumors can lead to a short average lifespan in their most severe form. If the brain tumor is found and detected early, the patient may have a better chance of survival. When using medical imaging, detecting a brain tumor is notoriously challenging. Factors such as the tumor’s size, shape, and location might vary from patient to patient. A tumor’s location in the brain makes it a difficult task to identify, therefore knowing these details are critical. It’s possible that some people have tumors of high glioma type, while others have tumors of low glioma type. As a result, it is vitally important to learn how to interpret medical imaging to identify a malignant tumor. Tumor and non-tumor images in medical imaging can be differentiated using a variety of machine learning methods. In this work we have used HOG to extract features from the medical data of patients with cancer tumors and then used them to classify using machine learning algorithms.