A malignant tumor in the brain is a life-threatening condition. It is known as glioblastoma, it's the‎ most common form of brain cancer in adults and the one with the worst prognosis, with a median survival of less than a year. This article is an extensive review of the basic background, technique and clinical applications of artificial intelligence (AI) and radiomics in the field of tumor neuroclassification. This survey provides a comprehensive review of recent advancements in brain tumor detection and MGMT promoter methylation status prediction using machine learning, deep learning, radiomic, and topological data analysis techniques. It encompasses a wide array of studies that utilize diverse datasets, such as TCGA, BraTS, RSNA, and Kaggle, showcasing the evolution of non-invasive diagnostic approaches. The reviewed research demonstrates significant improvements in diagnostic accuracy, often exceeding 90%, highlighting the potential of these methodologies to enhance early detection, classification, and personalized treatment planning in neuro-oncology. The survey emphasizes the importance of leveraging large, heterogeneous datasets and advanced algorithms to bridge the gap between research and clinical application. Future directions include validation across broader populations and integration into clinical workflows to realize the full potential of AI-driven brain tumor diagnostics.