Image segmentation of cervical vertebra in X-ray radiographs using the curve fitting strategy

Authors:
Huaifei Hu;Hong Liu;Lili Chen;Chih-Cheng Hung;Xiangyang Xu;Zhicong Lan
Affiliations:
Huazhong University of Science and Technology, Wuhan, China;Huazhong University of Science and Technology, Wuhan, China;Huazhong University of Science and Technology, Wuhan, China;State University, Marietta, GA;Huazhong University of Science and Technology, Wuhan, China;Huazhong University of Science and Technology, Wuhan, China
Venue:
Proceedings of the 2011 ACM Symposium on Applied Computing
Year:
2011

Citing 7
Cited 1

Active shape models—their training and application

Computer Vision and Image Understanding
Global Minimum for Active Contour Models: A Minimal Path Approach

International Journal of Computer Vision
User-steered image segmentation paradigms: live wire and live lane

Graphical Models and Image Processing
Fast Marching Methods

SIAM Review
Geodesic Active Contours and Level Sets for the Detection and Tracking of Moving Objects

IEEE Transactions on Pattern Analysis and Machine Intelligence
Customized Hough Transform for Robust Segmentation of Cervical Vertebrae from X-Ray Images

SSIAI '02 Proceedings of the Fifth IEEE Southwest Symposium on Image Analysis and Interpretation
Realization of the Contrast Limited Adaptive Histogram Equalization (CLAHE) for Real-Time Image Enhancement

Journal of VLSI Signal Processing Systems

An application of circumscribed circle filter in the Multi-Stencils Fast Marching method

Proceedings of the 27th Annual ACM Symposium on Applied Computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

We develop an effective method for the study of cervical vertebra maturation (CVM) for bone age evaluation. Such studies need an accurate X-ray radiographs segmentation of cervical vertebra. It is difficult to have a good segmentation on this type of images. Current segmentation methods do not work well on scanned images from analog image X-ray radiographs of cervical vertebra. A new method for analysis of cervical bone age is proposed. Two key techniques are developed in this proposed segmentation algorithm: (1) a fitting weight matrix is built to reduce the effect of subjective factors entered by the user when fast marching method is used to obtain the initial rough outline of cervical vertebra, and (2) apply a curve fitting method based on rotating and overlapping parabolic curves to derive the final segments of cervical vertebra. Furthermore, the user can calculate corresponding parameters from segmented results to assess the bone age. Experimental results using the proposed algorithm show that our algorithm is more accurate than those of fast marching method (FMM) and radiologists through repetition. It also shows that the proposed method has a higher accuracy on the correlation of the skeletal maturity indicators (SMI) and quantitative cervical vertebral maturation (QCVM).