eprintid: 3658
rev_number: 8
eprint_status: archive
userid: 366
dir: disk0/00/00/36/58
datestamp: 2019-11-28 08:59:04
lastmod: 2019-11-28 08:59:04
status_changed: 2019-11-28 08:59:04
type: conference_item
metadata_visibility: show
creators_name: Hoang, Hong Son
creators_name: Pham, Cam Phuong
creators_name: Franklin, Daniel
creators_name: Walsum, Theo van
creators_name: Luu, Manh Ha
creators_id: halm@vnu.edu.vn
title: An Evaluation of CNN-based Liver Segmentation
Methods using Multi-types of CT Abdominal
Images from Multiple Medical Centers
ispublished: pub
subjects: ElectronicsandComputerEngineering
divisions: avitech
divisions: fac_fet
abstract: Automatic segmentation of CT images has recently
been applied in several clinical liver applications. Convolutional Neural Networks (CNNs) have shown their effectiveness in medical image segmentation in general and also in liver segmentation. However, liver image quality may vary between medical centers due to differences in the use of CT scanners, protocols, radiation dose, and contrast enhancement. In this paper, we investigate three wells known CNNs, FCN-CRF, DRIU, and V-net, for liver segmentation using data from several medical centers. We perform qualitative evaluation of the CNNs based on Dice score, Hausdorff distance, mean surface distance and false positive rate. The results show that all three CNNs achieved a mean Dice score of over 90% in liver segmentation with typical contrast
enhanced CT images of the liver. p-values from paired T-test on Dice score of the three networks using Mayo dataset are larger than 0.05 suggesting that no statistical significant difference in their performance. DRIU performs the best in term of processing time. The results also demonstrate that those CNNs have reduced performance in liver segmentation in the case of low-dose and non-contrast enhanced CT images. In conclusion, these promising results enable further investigation of alternative deep learning based approaches to liver segmentation using CT images.
date: 2019-09-25
date_type: published
official_url: https://ieeexplore.ieee.org/document/8905166
full_text_status: none
pres_type: paper
event_title: 2019 19th International Symposium on Communications and Information Technologies (ISCIT)
event_location: Ho Chi Minh city
event_dates: 25-27/09/2019
event_type: conference
refereed: TRUE
citation:   Hoang, Hong Son and Pham, Cam Phuong and Franklin, Daniel and Walsum, Theo van and Luu, Manh Ha  (2019) An Evaluation of CNN-based Liver Segmentation Methods using Multi-types of CT Abdominal Images from Multiple Medical Centers.  In: 2019 19th International Symposium on Communications and Information Technologies (ISCIT), 25-27/09/2019, Ho Chi Minh city.