In recent years, medical images have established broad applications in healthcare for disease diagnosis, treatment planning, and disease progression control, which include the processing of images of the affected organ through different methods. Image segmentation is a process of partitioning an image into multiple segments used in different medical imaging applications. Aside from other organs, the accurate segmentation of MRI brain images allows the physician in the successful, early diagnosis and detection of brain tumors. The MRI brain images are susceptible to inhomogeneous noise intensities that inevitably influence pixel intensity. This, in turn, affects the accurate detection of tumor cells leading to incorrect segmentation results. It is known from the literature that current segmentation algorithms are usually dependent on the uniformity of the image intensities and therefore fail to deliver correct segmentation results in MR brain images that often contribute to a false position of tumor cells. This paper approach has explicitly revealed the procedures and methods to be followed after a detailed literature survey in segmenting the brain images affected by tumor. This paper focuses on developing segmentation algorithms for MRI brain images to diagnose tumor in the earlier stages by overcoming challenges in real time. This work has explicitly revealed the procedures and methodologies to be followed after a detailed literature survey in segmenting the brain images affected by tumor.Rather than keeping to a distinct segmentation standard, this research attempts to incorporate the benefits of different strategies to fit the criteria of segmentation of brain images. These algorithms are developed with the aim of applying to any kind of brain image that is intensely affected by tumor.
Volume 12 | Issue 6
Pages: 1915-1924
DOI: 10.5373/JARDCS/V12I2/S20201396