Segmentation Of Cartilage Tissue In Micro Ct Images Of Mouse Embryos With Modified U-Net Convolutional Neural Network

Matula, Jan

Segmentation Of Cartilage Tissue In Micro Ct Images Of Mouse Embryos With Modified U-Net Convolutional Neural Network

Files

191_eeict2019.pdf(667.56 KB)

Date

2019

Authors

Matula, Jan

Publisher

Vysoké učení technické v Brně, Fakulta elektrotechniky a komunikačních technologií

Abstract

Manual segmentation of cartilage tissue in micro CT images of mouse embryos is a very time-consuming process and significantly increases the time required for the research of mammal facial structure development. It is possible to solve this problem by using a fully-automatic segmentation algorithm. In this paper, a fully-automatic segmentation method is proposed using a convolutional neural network trained on manually segmented data. The architecture of the proposed convolutional network is based on the U-Net architecture with its encoding part substituted for the encoding part of the VGG16 classification convolutional neural network pre-trained on the ImageNet database of labelled images. The proposed network achieves average Dice coefficient 0.88 in comparison to manually segmented images.

Keywords

segmentation, cartilage, convolutional neural networks, deep learning

Citation

Proceedings of the 25st Conference STUDENT EEICT 2019. s. 191-194. ISBN 978-80-214-5735-5
http://www.feec.vutbr.cz/EEICT/

Document type

Peer-reviewed

Document version

Published version

Language of document

en

Document licence