Recent Trends in Brain Tumor Prediction Using Machine Learning and Deep Learning

A spontaneous growth of abnormal cells in brain tissues that increases the pressure inside the skull is called a brain tumor. This tumor can be non-cancerous or cancerous based on characteristics, but at most can cause life-threatening if not treated in the early stages. Machine learning and deep learning approaches have become a potent tool in the field of medical imaging in recent years, providing brain tumor identification and classification with increased automation and accuracy. This review explores the recent advancements in machine learning and deep learning models, which have shown better results in brain tumor prediction. Mainly the type of images that are considered for brain tumor capturing are Magnetic Resonance Imaging (MRI), as these images provide great insights into brain tissues, so in this paper, we are referring to MRI-type images only. This paper also covers various segmentation techniques mentioned in multiple studies, as it is a time-consuming job for radiologists to deeply understand MRI images and highlight the tumor present. In this paper, we have also mentioned what metrics are used to analyze these models, like accuracy, sensitivity, specificity, and more, to know if the models are robust.