International Journal of Embedded Systems and Emerging Technologies

Deep Learning Approach Based Brain Tumor Detection

  1. Kamlesh Kumar singh
  2. Bramha Hazela
  3. Deependra Pandey
  4. Akanksha Kumari

Abstract

The brain tumor is one of the most deadly disease which is of great concern and hence to detect and extract the effected tumor area in the brain from the MRI has lot of research potential. It is also a very difficult task to perform by the experts. The perfection of the vivisection depends on how much experienced the radiologist or the clinical expert is. So to overcome the wastage of time and to overcome the limitation of the presence of experienced clinical expert we can use the concept of image processing and the practical implementation of deep learning. Here, in this research, we have used the CNN model of deep learning for the segmentation of magnetic resonance images. An experimental result of 90.42% was achieved by using the CNN approach. By this approach 90.42 % accuracy identifying between the MR images consisting tumors in the brain and the ones which do not have the tumor cells. The results  showed an average of 0.83 dice similarity index which clearly indicates the better overlap of the tumor region from the image processed model and the manual extraction by the clinical experts. By the results which we achieved we found that the level of prediction and segmentation of brain tumors are almost similar if we used computer-aided technology or if we go by the clinical approach. Hence we conclude that a computer-aided technology is less time taking and give a quick result with very less effort.

Read more of it

Subscribe to one journal or build your own list of several. One invoice, one payment, access for a full year.

Subscribe to journals →

Publish your article

Peer-reviewed journals across every subject, with open access options. Send your manuscript to the right editor today.

Publish with us → Calculate your APC
Support