Segmentation of Brain Tumor Tissues in HGG and LGG MR Images Using 3D U-net Convolutional Neural Network

Segmentation of Brain Tumor Tissues in HGG and LGG MR Images Using 3D U-net Convolutional Neural Network

Poornachandra Sandur, C. Naveena, V.N. Manjunath Aradhya, Nagasundara K. B.
Copyright: © 2018 |Pages: 13
DOI: 10.4018/IJNCR.2018040102
(Individual Articles)
No Current Special Offers


The quantitative assessment of tumor extent is necessary for surgical planning, as well as monitoring of tumor growth or shrinkage, and radiotherapy planning. For brain tumors, magnetic resonance imaging (MRI) is used as a standard for diagnosis and prognosis. Manually segmenting brain tumors from 3D MRI volumes is tedious and depends on inter and intra observer variability. In the clinical facilities, a reliable fully automatic brain tumor segmentation method is necessary for the accurate delineation of tumor sub regions. This article presents a 3D U-net Convolutional Neural Network for segmentation of a brain tumor. The proposed method achieves a mean dice score of 0.83, a specificity of 0.80 and a sensitivity of 0.81 for segmenting the whole tumor, and for the tumor core region a mean dice score of 0.76, a specificity of 0.79 and a sensitivity of 0.73. For the enhancing region, the mean dice score is 0.68, a specificity of 0.73 and a sensitivity of 0.77. From the experimental analysis, the proposed U-net model achieved considerably good results compared to the other segmentation models.
Article Preview

Literature Survey

This Section highlights the active researchers working in the area of brain tumor segmentation. The research, development, and scientific publications in the area of computerized brain tumor segmentation have grown faster in the last few decades, and this shows that the brain tumor segmentation research is still a work in progress but also signifies the need for computerized brain tumor segmentation methods in large clinical institutions.

Segmentation of brain tumor methods is grouped into two types: Generative model and the other is based on the Discriminative model. The Generative models employ detailed prior knowledge about the spatial distribution and appearance of cells. These models display good generalization to unseen images and has represented promising results for segmentation of brain tumors. The prior knowledge encoding of a brain tumor is a tedious task. Based on the expected shape or image signal of non-tumorous brain tissues, brain tumors are modelled as outliers. The tumor specific “bio-markers” are employed to extract the spatial information. And, also models of tumor growth are employed to infer the probable tumor structures. The Generative models depend on image registration for properly aligning the images and spatial priors; which is a hard task in presence of big size brain tumor structures.

Complete Article List

Search this Journal:
Volume 12: 1 Issue (2023): Forthcoming, Available for Pre-Order
Volume 11: 4 Issues (2022): 1 Released, 3 Forthcoming
Volume 10: 4 Issues (2021)
Volume 9: 4 Issues (2020)
Volume 8: 4 Issues (2019)
Volume 7: 4 Issues (2018)
Volume 6: 2 Issues (2017)
Volume 5: 4 Issues (2015)
Volume 4: 4 Issues (2014)
Volume 3: 4 Issues (2012)
Volume 2: 4 Issues (2011)
Volume 1: 4 Issues (2010)
View Complete Journal Contents Listing