Home About Login Current Archives Announcements Editorial Board
Submit Now For Authors Call for Submissions Statistics Contact
Home > Archives > Volume 20, No 8 (2022) > Article

DOI: 10.14704/nq.2022.20.8.NQ44425

Brain Tumor MRI Image Segmentation using Convolution Neural Network



A lump or development of abnormal cells in your brain is known as a brain tumour. There are several varieties of brain tumours. Some brain tumours are benign (noncancerous), while others are malignant (malignant). The greatest grade of glioma, the smallest brain tumour yet recognised, has an extremely poor prognosis. Thus, a crucial step in enhancing the quality of life for cancer patients is treatment planning. Multi-sequence MRI techniques for segmenting brain tumours are not standardised in clinical practise, necessitating the employment of a flexible segmentation strategy that makes the most use of all available MRI data. In order to precisely and successfully segment a tumour, we present an autonomous segmentation approach based on convolutional neural networks (CNN) that explores tiny kernels while also being effective against overfitting. Due to the smaller number of weights in the network, using tiny kernels enables the creation of a more complex architecture. To discover force markers from the training set, we employed depth normalisation as a pre-processing step. Later, supervised image classification using CNN is performed. Additionally, it accurately segments the tumour in the MRI picture.


Convolution Neural Networks, Tumor, MRI, Image Segmentation.

Full Text