Contour - Marker Based Segmentation For Tumorous And Non-Tumorous Brain Mri Detection

• Published in: 2021 10th International Conference on Internet of Everything, Microwave Engineering, Communication and Networks (IEMECON) pp. 01 - 06
• Main Authors: Chowdhury, Debkumar, Mishra, Sanjukta, Mondal, Sayantika, Singh, Anjali, Guha, Koustav, Aziz, Md. Tarik, Sen, Rahul, Chowdhury, Subhankar Bhanja, Porey, Sayan, Sau, Kartik
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.12.2021
• Subjects: Brain modeling; Brain Tumor; Computational modeling; Computer vision; Filtering; Image segmentation; Magnetic resonance imaging; marker-based image segmentation; Prediction algorithms; Predictive models
• DOI: 10.1109/IEMECON53809.2021.9689131

Brain Tumor is a serious concern and can be a cause of death if not diagnosed properly. The fatal rate can be avoided by early detection and treatment. Computer vision technique helps to analyze brain tumors automatically and effectively. In this paper, we propose a marker-based image segmentation technique to find the presence or absence of tumors in brain MRI images. We use a standard dataset to build the model. The model is based on the comparisons among different filtering, thresholding, and segmentation techniques to find out the best method which predicts the result with great accuracy. We evaluate the filtering methods using PSNR, SNR, MSE, RMSE Values. Finally, a marker-based algorithm has been used for detection. The Experimental result shows that our attempts are promising, and the model performs well in detecting the abnormal image.