Affiliations: [a] Department of Information Technology, Sathyabama Institute of Science and Technology, Chennai, India
| [b] Department of Information Technology, Jeppiaar SRR Engineering College, Padur, Chennai, India
| [c] Department of Computer Science and Engineering, Sathyabama Institute of Science and Technology, Chennai, India
Correspondence:
[*]
Corresponding author. A. Veeramuthu, Department of Information Technology, Sathyabama Institute of Science and Technology, Chennai. E-mail: aveeramuthu@gmail.com.
Abstract: Brain tumor image segmentation is process of locating the interesting area in terms of objects, like tumor and extracting it for the further process of the image and getting the boundaries of the image for analysis. The bio-medical brain tumor image segmentation is a great challenging field for the today world active researchers with the standardized image datasets and various metrics used for evaluating and comparing the performance of the new algorithm with existing segmentation algorithms. In recent development, these problems are addressed using various image manipulation tools and rapid growth of computer hardware enhancement. Image segmentation was done in three ways: (1) Manual-based (2) Semi-automated-based (3) Fully automated-based. But still be a short of research in the field of brain tumor segmentation and accurate identification of tumor cells. To overcome all the above-mentioned challenges and complexity of the brain tumor segmentation, it need to understand the pre-processing of the image like, registering the image, correction of bias in image, and non-brain tissue removal. In this paper, we propose a new methodology for segmenting the brain tumor from the affected brain image in a significantly efficient way by using deep learning method.
Keywords: Segmentation, metrics, manual-based, semi-automated, automated, deep learning
