Journal article Open Access

The Multi Stage U-net Design for Brain Tumor Segmentation using Deep Learning Architecture

Putta. Rama Krishna Veni,; C Aruna Bala

Blue Eyes Intelligence Engineering and Sciences Publication(BEIESP)

Now a day’s diagnosis and accurate segmentation of brain tumors are critical conditions for successful treatment. The manual segmentation takes time consuming process, more cost and inaccurate. In this paper implementation of cascaded U-net segmentation Architecture are divided into substructures of brain tumor segmentation. The neural network is competent of end to end multi modal brain tumor segmentations.The Brain tumor segments are divided three categories. The tumor core (TC),the enhancing tumor(ET),the whole tumor (WT).The distinct data enhancement steps are better achievement. The proposed method can test result conclude average counter scores of 0.83268, 0.88797 and 0.83698, as well as Hausdorff distances 95%) of 2.65056, 4.61809 and 4.13071, for the enhancing tumor(ET), whole tumor (WT) and tumor core (TC) respectively. In this method validating with BraTS 2019 dataset and identify the test time enhancement improves the Brain tumor segmentation accurate images.

Files (442.3 kB)
Name Size
442.3 kB Download
All versions This version
Views 2929
Downloads 2828
Data volume 12.4 MB12.4 MB
Unique views 2222
Unique downloads 2727


Cite as