A Comprehensive Survey of Machine Learning Techniques for Brain Tumor Detection

Cells in brain, which develop too quickly and uncontrollably, can lead to brain tumor. Early detection and treatment are crucial for the treatment. However, accurately segmenting and categorizing tumor remains a daunting challenge despite the numerous research efforts. CNNs, a well-liked deep learning (DL) model, may nevertheless be constrained by major input variances or domain shifts. Pre-trained transfer-based learning models are becoming more and more common as a solution to this issue. This survey aims to present a thorough overview of shift from state-of-the-art CNN-based models to transfer learning-based techniques, to automate brain tumor detection. This review discusses the reason of shift along with architecture of brain tumor, publicly accessible datasets, segmentation, feature extraction, classification, and cutting-edge approaches for studying brain cancers including machine learning, deep learning, and transfer learning. In addition, relevant issues and typical difficulties have also been highlighted.