Dental segmentation from X-ray images using semi-supervised fuzzy clustering with spatial constraints

In this paper, we propose a novel semi-supervised fuzzy clustering algorithm with spatial constraints for dental segmentation from X-ray images. The detailed contributions include: i) Formulating the spatial features of a dental X-ray image in a dental feature database; ii) Modeling the dental segmentation problem in the form of semi-supervised fuzzy clustering with spatial constraints; iii) Solving the model by the Lagrange multiplier method; iv) Determining the additional information for clustering process by mixing optimal results of Fuzzy C-Means with spatial constraints; v) Proposing a novel Semi-Supervised Fuzzy Clustering algorithm with Spatial Constraints (SSFC-SC) that combines those processes for dental segmentation. The new algorithm is validated on a real dataset from Hanoi Medical University, Vietnam including 56 dental images. The experimental results reveal that the proposed work has better accuracy than the original semi-supervised fuzzy clustering and other relevant methods. We also suggest the most appropriate values of parameters that should be opted for the algorithm.

Title:	Dental segmentation from X-ray images using semi-supervised fuzzy clustering with spatial constraints
Authors:	Le Hoang Son Tran Manh Tuan
Keywords:	Dental features;Dental image segmentation;Fuzzy clustering;Semi-supervised fuzzy clustering;X-ray images
Issue Date:	2017
Publisher:	PERGAMON-ELSEVIER SCIENCE LTD, THE BOULEVARD, LANGFORD LANE, KIDLINGTON, OXFORD OX5 1GB, ENGLAND
Citation:	ISIKNOWLEDGE
Abstract:	In this paper, we propose a novel semi-supervised fuzzy clustering algorithm with spatial constraints for dental segmentation from X-ray images. The detailed contributions include: i) Formulating the spatial features of a dental X-ray image in a dental feature database; ii) Modeling the dental segmentation problem in the form of semi-supervised fuzzy clustering with spatial constraints; iii) Solving the model by the Lagrange multiplier method; iv) Determining the additional information for clustering process by mixing optimal results of Fuzzy C-Means with spatial constraints; v) Proposing a novel Semi-Supervised Fuzzy Clustering algorithm with Spatial Constraints (SSFC-SC) that combines those processes for dental segmentation. The new algorithm is validated on a real dataset from Hanoi Medical University, Vietnam including 56 dental images. The experimental results reveal that the proposed work has better accuracy than the original semi-supervised fuzzy clustering and other relevant methods. We also suggest the most appropriate values of parameters that should be opted for the algorithm
Description:	TNS06979 ; ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE Volume: 59 Pages: 186-195 Published: MAR 2017
URI:	http://repository.vnu.edu.vn/handle/VNU_123/28304 http://www.sciencedirect.com/science/article/pii/S0952197617300039
ISSN:	0952-1976 1873-6769
Appears in Collections:	Bài báo của ĐHQGHN trong Web of Science