Brain-Tumor Detection And Segmentation Using Machine Learning Techniques

Abstract

The most significant part of the human body is the brain. It regulates and plans how we act and communicate. The complexity of the brain's architecture is a significant hurdle in necessitating prompt and correct diagnosis. Early diagnosis improves survival prospects and treatment options. In order to recognize and diagnose brain cancers earlier, Artificial intelligence is playing a crucial rule. Recent advances in AI's machine-learning and deep-learning have completely changed how neurosurgical procedures are performed. These include feature-extraction, feature-selection, feature-reduction, classification, data enrichment, and data preprocessing. The research publications on the segmentation and detection of brain-tumors using magnetic resonance imaging (MRI)-images from the recent past are reviewed in this article. Each research paper's fundamental segmentation methods were carefully reviewed. This paper offers a comprehensive review of the subject as well as fresh perspectives on the various machine-learning and image segmentation techniques used to detect brain-tumors. Deep-learning approaches are more efficient for segmenting and detecting the tumor from brain MRI images, and data augmentation techniques can improve the performance of the tumor identification process

Keywords: Machine-learning, Convolutional-Neural-Network CNN, Deep-Learning, Data Augmentation.