Title: Variability of Cobb angle measurement from digital X-ray image based on different de-noising techniques

Authors: Jhilam Mukherjee; Raka Kundu; Amlan Chakrabarti

Addresses: A.K.Choudhury School of Information Technology, University of Calcutta, Kolkata, West Bengal, India ' A.K.Choudhury School of Information Technology, University of Calcutta, Kolkata, West Bengal, India ' A.K.Choudhury School of Information Technology, University of Calcutta, Kolkata, West Bengal, India

Abstract: Scoliosis is a 3D deformity of the spine, which is characterised by both lateral curvature and vertebral rotation. Cobb angle measurement is the golden method for diagnosis of scoliosis. With the increasing trend of digital imaging-based diagnosis system, computerised Cobb angle measurement is gaining much popularity. In general, digital X-ray images are very sensitive to noise, so a good pre-processing step for noise removal is necessary to increase the accuracy and reduce the variability of computerised Cobb angle measurement method. In this paper, we have presented a review work on computerised Cobb angle measurement from digital X-ray images based on the different popular state-of-the-art de-noising techniques. This study helps in selecting the de-noising technique that best suits with different steps of the Cobb angle method and generates a more reliable angular measurement. This type of study based on de-noising technique for reducing Cobb angle measurement variability is the first of its kind. MATLAB2011Rb image processing toolbox was used for the simulation and verification of our proposed methodology.

Keywords: Cobb angle measurement; PND-NLM; BM3D; non-local means; bilateral filters; digital images; X-ray image; X-rays; variability; de-noising; scoliosis diagnosis; spine deformity; medical imaging; simulation.

DOI: 10.1504/IJBET.2014.065656

International Journal of Biomedical Engineering and Technology, 2014 Vol.16 No.2, pp.113 - 134

Received: 27 Dec 2013
Accepted: 07 Jul 2014
Published online: 25 Apr 2015 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article