Segmentation of Low-Grade Gliomas using Deep U-Net with Transfer Learning

Rasyid, Dwilaksana Abdullah (2021) Segmentation of Low-Grade Gliomas using Deep U-Net with Transfer Learning. Masters thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 06211850012003-Master_Thesis.pdf] Text
06211850012003-Master_Thesis.pdf - Accepted Version
Restricted to Repository staff only

Download (1MB) | Request a copy


The World Health Organization (WHO) grades II and III gliomas or low-grade gliomas (LGG) are slow-growing brain tumors. LGG is a fatal disease of young adults (between 35 and 44 years of age). LGG can transform into high-grade gliomas or WHO grade IV gliomas, which occurs in most patients and ultimately leads to death. General treatment for LGG patients is surgical resection, radiotherapy, and chemotherapy. Fluid-attenuated inversion recovery (FLAIR) imaging is needed to determine the tumor location before doing surgical resection. We propose an architectural innovation of combining U-Net and VGG16 with transfer learning for LGG tumor segmentation. Employing the preoperative FLAIR imaging data of 110 patients with LGG from the Cancer Genome Atlas, this deep learning algorithm achieves an excellent result with the Dice similarity coefficient of 99% and the area under the curve of 98%, better than the previous approach done by Buda, et al.

Item Type: Thesis (Masters)
Uncontrolled Keywords: Low-Grade Gliomas, Brain Tumor Segmentation, U-Net, VGG16, Transfer Learning, Deep Learning
Subjects: Q Science > Q Science (General)
Q Science > Q Science (General) > Q325.5 Machine learning.
Q Science > QH Biology
Divisions: Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49101-(S2) Master Thesis
Depositing User: Dwilaksana Abdullah Rasyid
Date Deposited: 10 Mar 2021 04:03
Last Modified: 10 Mar 2021 04:03

Actions (login required)

View Item View Item