Brain Tumor Detection and Segmentation Using Mask R-CNN
Haidarali Tajuddin Mulla
, Atish Vijay Kamble , Shreyas Satish Mahamuni , Suraj Ravindra Khare , Mrs. S. J. Chougule
Abstract, Introduction, Litrature Survey, Methodology, Result, Hardware & Software used in proposed system, Conclusion, References.
One of the dreadful diseases that the world encounters moment is brain tumors. When abnormal cells form in the brain, it is called a brain tumor. There are a lot of variations in the sizes and positions of tumors, and hence this makes it really hard for a complete understanding of tumours. Radiologists can fluently diagnose the disease with the help of medical image techniques, but making this process automatic is obviously useful. Magnetic Resonance Imaging (MRI) is the most effective system for detecting brain tutors where, MRI images are trained and tested in order to descry the tumor. The automated system would be suitable to find and pinpoint the exact position of the tumor in an MRI image. Our study describes a method for segmenting abnormal brain tissues and determining whether the case has a tumor. This approach detects a unique area of the brain and forecasts the liability of a tumor developing there. Mask regional-based convolution neural network (Mask R-CNN) is a pre-trained deep neural network model that is used to distinguish objects from an image such as buses, animals, persons, trees, and other objects. In comparison to numerous other analogous methods based on MLP, VGG-16 model, and U-net model, we discovered that Mask R-CNN method performs the best. The clarity of the MRI scans has a big impact on the delicacy. The proposed system was suitable to outperform similar systems on the same dataset, achieving a 74 percent crossroad over Union (IoU) score on the reference dataset, Brain MRI Images for Brain Tumor Detection. The demand for effective computer-aided brain tumour segmentation techniques has increased vastly in recent times. Still, accurate brain tumour segmentation is still a challenge because of its structural complexity such as variations in position, size, shape, overlapping tumor boundaries with normal brain tissues, etc. Existing automated approaches for brain tumour detection can be broadly categorized into handwrought features and deep learning (DL) based approaches. Qasem et al. [1] used a watershed segmentation algorithm along with the KNN for brain tumour classification and segmentation. This method performs well on the selected MRI images and is unable to accurately segment the tumour regions on challenging images containing tumors with multiple structural complexities
"Brain Tumor Detection and Segmentation Using Mask R-CNN", IJSDR - International Journal of Scientific Development and Research (www.IJSDR.org), ISSN:2455-2631, Vol.8, Issue 3, page no.407 - 411, March-2023, Available :https://ijsdr.org/papers/IJSDR2303063.pdf
Volume 8
Issue 3,
March-2023
Pages : 407 - 411
Paper Reg. ID: IJSDR_204412
Published Paper Id: IJSDR2303063
Downloads: 000347191
Research Area: Engineering
Country: Sangli, Maharashtra, India
ISSN: 2455-2631 | IMPACT FACTOR: 9.15 Calculated By Google Scholar | ESTD YEAR: 2016
An International Scholarly Open Access Journal, Peer-Reviewed, Refereed Journal Impact Factor 9.15 Calculate by Google Scholar and Semantic Scholar | AI-Powered Research Tool, Multidisciplinary, Monthly, Multilanguage Journal Indexing in All Major Database & Metadata, Citation Generator
Publisher: IJSDR(IJ Publication) Janvi Wave
