A proper diagnosis and effective treatment of a brain tumour require trustworthy segmentation. Automated solutions for brain tumour segmentation are often welcomed due to the high cost, long duration, and inherent subjectivity of the traditional process. However, developing automatic segmentation algorithms for these tumours has been a difficult task for the past few decades due to the location-, shape-, and size-specific heterogeneity of brain tumors. In this paper, we develop a multi-model deep learning segmentation of brain tumor images. The model is developed in such a way that it segments well the regions of tumour regions. The simulation is conducted to find the model efficacy. The results of simulation shows that the proposed method achieves higher grade of segmentation accuracy than the other existing methods.

brain tumour, segmentation, multi-model segmentation, deep learning