A systematic review on deep learning based brain tumor segmentation and detection using MRI: Past insights, present techniques and future trends.

Abstract

The abnormal growth of cells leads to brain malignancy in humans, which is among the most prevalent causes of fatalities in adults worldwide. Patients' likelihood of survival increases, and therapeutic opportunities improve when brain tumors are identified early. Compared to other imaging techniques, Magnetic Resonance Imaging (MRI) scans provide more comprehensive information. A brain tumor can be diagnosed and differentiated from MRI images using a variety of brain tumor recognition and segmentation approaches. The utilization of deep learning-based models has proven effective in analyzing the vast volume of MRI data. The main purpose of this review is to provide an overview of brain tumor segmentation and detection techniques. To efficiently process the large volume of images, this review presents a detailed analysis of deep learning models. Furthermore, a chronological analysis is carried out to validate the robustness of the techniques. Following that, to better understand the performance of the models, the strengths and limitations of standard deep learning methods are discussed. In addition, the dataset details, performance evaluations, and simulation tools are discussed in this review. Finally, the challenges and research gaps in brain tumor segmentation and detection models are highlighted.