Staff View: Unifying the seeds auto generation (SAGE) with knee cartilage segmentation framework: data from the osteoarthritis initiative

Unifying the seeds auto generation (SAGE) with knee cartilage segmentation framework: data from the osteoarthritis initiative

Purpose: Manual segmentation is sensitive to operator bias, while semiautomatic random walks segmentation offers an intuitive approach to understand the user knowledge at the expense of large amount of user input. In this paper, we propose a novel random walks seed auto-generation (SAGE) hybrid mode...

Full description

Saved in:

Bibliographic Details
Main Authors:	Gan, Hong Seng, Sayuti, Khairil Amir, Ramlee, Muhammad Hanif, Lee, Yeng Seng, Wan Mahmud, Wan Mahani Hafizah, Abdul Karim, Ahmad Helmy
Format:	Article
Published:	Springer Verlag 2019
Subjects:	QH Natural history TP Chemical technology
Online Access:	http://eprints.utm.my/id/eprint/89236/ http://dx.doi.org/10.1007/s11548-019-01936-y
Tags:	Add Tag No Tags, Be the first to tag this record!

id	my.utm.89236
record_format	eprints
spelling	my.utm.892362021-02-22T06:01:13Z http://eprints.utm.my/id/eprint/89236/ Unifying the seeds auto generation (SAGE) with knee cartilage segmentation framework: data from the osteoarthritis initiative Gan, Hong Seng Sayuti, Khairil Amir Ramlee, Muhammad Hanif Lee, Yeng Seng Wan Mahmud, Wan Mahani Hafizah Abdul Karim, Ahmad Helmy QH Natural history TP Chemical technology Purpose: Manual segmentation is sensitive to operator bias, while semiautomatic random walks segmentation offers an intuitive approach to understand the user knowledge at the expense of large amount of user input. In this paper, we propose a novel random walks seed auto-generation (SAGE) hybrid model that is robust to interobserver error and intensive user intervention. Methods: Knee image is first oversegmented to produce homogeneous superpixels. Then, a ranking model is developed to rank the superpixels according to their affinities to standard priors, wherein background superpixels would have lower ranking values. Finally, seed labels are generated on the background superpixel using Fuzzy C-Means method. Results: SAGE has achieved better interobserver DSCs of 0.94 ± 0.029 and 0.93 ± 0.035 in healthy and OA knee segmentation, respectively. Good segmentation performance has been reported in femoral (Healthy: 0.94 ± 0.036 and OA: 0.93 ± 0.034), tibial (Healthy: 0.91 ± 0.079 and OA: 0.88 ± 0.095) and patellar (Healthy: 0.88 ± 0.10 and OA: 0.84 ± 0.094) cartilage segmentation. Besides, SAGE has demonstrated greater mean readers’ time of 80 ± 19 s and 80 ± 27 s in healthy and OA knee segmentation, respectively. Conclusions: SAGE enhances the efficiency of segmentation process and attains satisfactory segmentation performance compared to manual and random walks segmentation. Future works should validate SAGE on progressive image data cohort using OA biomarkers. Springer Verlag 2019-03 Article PeerReviewed Gan, Hong Seng and Sayuti, Khairil Amir and Ramlee, Muhammad Hanif and Lee, Yeng Seng and Wan Mahmud, Wan Mahani Hafizah and Abdul Karim, Ahmad Helmy (2019) Unifying the seeds auto generation (SAGE) with knee cartilage segmentation framework: data from the osteoarthritis initiative. International Journal of Computer Assisted Radiology and Surgery, 14 . pp. 755-762. ISSN 1861-6429 http://dx.doi.org/10.1007/s11548-019-01936-y DOI:10.1007/s11548-019-01936-y
institution	Universiti Teknologi Malaysia
building	UTM Library
collection	Institutional Repository
continent	Asia
country	Malaysia
content_provider	Universiti Teknologi Malaysia
content_source	UTM Institutional Repository
url_provider	http://eprints.utm.my/
topic	QH Natural history TP Chemical technology
spellingShingle	QH Natural history TP Chemical technology Gan, Hong Seng Sayuti, Khairil Amir Ramlee, Muhammad Hanif Lee, Yeng Seng Wan Mahmud, Wan Mahani Hafizah Abdul Karim, Ahmad Helmy Unifying the seeds auto generation (SAGE) with knee cartilage segmentation framework: data from the osteoarthritis initiative
description	Purpose: Manual segmentation is sensitive to operator bias, while semiautomatic random walks segmentation offers an intuitive approach to understand the user knowledge at the expense of large amount of user input. In this paper, we propose a novel random walks seed auto-generation (SAGE) hybrid model that is robust to interobserver error and intensive user intervention. Methods: Knee image is first oversegmented to produce homogeneous superpixels. Then, a ranking model is developed to rank the superpixels according to their affinities to standard priors, wherein background superpixels would have lower ranking values. Finally, seed labels are generated on the background superpixel using Fuzzy C-Means method. Results: SAGE has achieved better interobserver DSCs of 0.94 ± 0.029 and 0.93 ± 0.035 in healthy and OA knee segmentation, respectively. Good segmentation performance has been reported in femoral (Healthy: 0.94 ± 0.036 and OA: 0.93 ± 0.034), tibial (Healthy: 0.91 ± 0.079 and OA: 0.88 ± 0.095) and patellar (Healthy: 0.88 ± 0.10 and OA: 0.84 ± 0.094) cartilage segmentation. Besides, SAGE has demonstrated greater mean readers’ time of 80 ± 19 s and 80 ± 27 s in healthy and OA knee segmentation, respectively. Conclusions: SAGE enhances the efficiency of segmentation process and attains satisfactory segmentation performance compared to manual and random walks segmentation. Future works should validate SAGE on progressive image data cohort using OA biomarkers.
format	Article
author	Gan, Hong Seng Sayuti, Khairil Amir Ramlee, Muhammad Hanif Lee, Yeng Seng Wan Mahmud, Wan Mahani Hafizah Abdul Karim, Ahmad Helmy
author_facet	Gan, Hong Seng Sayuti, Khairil Amir Ramlee, Muhammad Hanif Lee, Yeng Seng Wan Mahmud, Wan Mahani Hafizah Abdul Karim, Ahmad Helmy
author_sort	Gan, Hong Seng
title	Unifying the seeds auto generation (SAGE) with knee cartilage segmentation framework: data from the osteoarthritis initiative
title_short	Unifying the seeds auto generation (SAGE) with knee cartilage segmentation framework: data from the osteoarthritis initiative
title_full	Unifying the seeds auto generation (SAGE) with knee cartilage segmentation framework: data from the osteoarthritis initiative
title_fullStr	Unifying the seeds auto generation (SAGE) with knee cartilage segmentation framework: data from the osteoarthritis initiative
title_full_unstemmed	Unifying the seeds auto generation (SAGE) with knee cartilage segmentation framework: data from the osteoarthritis initiative
title_sort	unifying the seeds auto generation (sage) with knee cartilage segmentation framework: data from the osteoarthritis initiative
publisher	Springer Verlag
publishDate	2019
url	http://eprints.utm.my/id/eprint/89236/ http://dx.doi.org/10.1007/s11548-019-01936-y
_version_	1692991762863226880
score	13.211869

Unifying the seeds auto generation (SAGE) with knee cartilage segmentation framework: data from the osteoarthritis initiative

Similar Items