| Abstrakt: |
Detecting tumour with Magnetic Resonance imaging (MRI) has been promising in recent years. The current tumour detection strategies are used for visualising, digitising and modelling the images for medical diagnosis. Automatic tumour region segmentation from 3D MRIs is essential for monitoring, diagnosing, and planning the treatment remedies for the disease. However, manual outlining needs an anatomical understanding which requires more time and can be imprecise due to human error. Brain tumour is serious diseases whose exposure must be quick and precise. It is attained by the accomplishment of the automatic detection of the tumour using MRI. Several automatic strategies are employed in literary works, which utilise image segmentation. Here, 50 research papers are surveyed with brain tumour detection strategies like ANN, clustering, deep learning, edge-based techniques, optimisation and transformation strategies. Furthermore, complete exploration is done in terms of publication year, employed strategy, datasets utilised, execution tool, performance measures and its values. At last, the problems of classical strategies and its issues with classical brain tumour detection methods are illustrated to generate the contribution. [ABSTRACT FROM AUTHOR] |
