Brain Tumour Detection Through Modified UNet-Based Semantic Segmentation

Abstract

The determination of the tumor's extent is a major challenge in brain tumour treatment planning and measurement. Non-invasive magnetic resonance imaging (MRI) has evolved as a first-line diagnostic tool for brain malignancies without the use of ionising radiation. Manually segmenting the extent of a brain tumour from 3D MRI volumes is a time-consuming process that significantly relies on the experience of the operator. As a result, we suggested a modified UNet structure based on residual networks that use periodic shuffling at the encoder region of the original UNet and sub-pixel convolution at the decoder section in this research. The proposed UNet was tested on BraTS Challenge 2017 with high-grade glioma (HGG). The model was tested on BraTS 2017 and 2018 datasets. Tumour core (TC), whole tumour (WT), and enhancing core (EC) were the three major labels to be segmented. The test results shown that proposed UNet outperform the existing techniques.