Texture Feature Extraction from Thyroid MR Imaging Using High-Order Derived Mean CLBP

• Published in: Journal of computer science and technology Vol. 34; no. 1; pp. 35 - 46
• Main Authors: Liu, Zhe, Qiu, Cheng-Jian, Song, Yu-Qing, Liu, Xiao-Hong, Wang, Juan, Sheng, Victor S.
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.01.2019; Springer; Springer Nature B.V; School of Computer Science and Telecommunication, Jiangsu University, Zhenjiang 212013, China%Department of Computer Science, University of Central Arkansas, Arkansas 72035, U.S.A
• Subjects: Algorithms; Area classification; Artificial Intelligence; Coding; Computer Science; Data Structures and Information Theory; Feature extraction; Image classification; Information Systems Applications (incl.Internet); Magnetic resonance imaging; Medical imaging; Medical imaging equipment; Noise sensitivity; Pixels; Regular Paper; Software Engineering; Texture; Theory of Computation; Thyroid gland; texture feature; local binary pattern; thyroid magnetic resonance imaging (MRI); complete local binary pattern (CLBP)
• ISSN: 1000-9000; 1860-4749
• DOI: 10.1007/s11390-019-1897-9

In the field of medical imaging, the traditional local binary pattern (LBP) and its improved algorithms are often sensitive to noise. Traditional LBPs are solely based on the signal information from local differences, and the binary quantization method oversimplifies the local texture features while disregarding the imaging information from the concaveconvex regions between the high-order pixels and the neighboring sampling points. Therefore, we propose an improved Derived Mean Complete Local Binary Pattern (DM CLBP) algorithm based on high-order derivatives. In the DM CLBP method, the grey value of a single pixel is replaced by the mean grey value of the rectangular area block, and the difference between pixel values in the area is obtained using the second-order differentiation method. Based on the calculation concept of the complete local binary pattern (CLBP) algorithm, the cascade signs and magnitudes of the two components are encoded and recombined in DM CLBP using a uniform pattern. The results from the experiments showed that the proposed DM CLBP descriptors achieved a classification accuracy of 94.4%. Compared with LBP and other improved algorithms, the DM CLBP algorithm presented in this study can effectively differentiate between lesion areas and normal areas in thyroid MR (magnetic resonance) images and shows the improved accuracy of area classification.