Diagnosis of Thyroid Nodules Based on Local Non-quantitative Multi-Directional Texture Descriptor with Rotation Invariant Characteristics for Ultrasound Image
- 9 Downloads
The traditional texture feature lacks the directional analysis of graphical element, so it could not better distinguish the thyroid nodule texture image formed by the rotation of graphical element. A non-quantifiable local feature is adopted in this paper to design a robust texture descriptor so as to enhance the robustness of the texture classification in the rotation and scale changes, which can improve the diagnostic accuracy of thyroid nodules in ultrasound images. First of all, the concept of local feature with rotational symmetry is introduced. It is found that many rotation invariant local features are rotational symmetric to a certain degree. Therefore, we propose a novel local feature to describe the rotation invariant properties of the texture. In order to deal with the change of rotation and scale of ultrasound thyroid nodules in image, Pairwise rotation-invariant spatial context feature is adopted to analyze the texture feature, which can combine with the scale information without increasing the dimension of the local feature. The fadopted local features have strong robustness to rotation and gray intensity variation. The experimental results show that our proposed method outperforms the existing algorithms on thyroid ultrasound data sets, which greatly improve the Diagnosis accuracy of thyroid nodules.
KeywordsThyroid nodule Texture descriptor Non-quantitative Multi-direction SVM classifier Rotation invariant Local feature
Compliance with Ethical Standards
Conflict of Interest
We declare that we have no conflict of interest. The paper does not contain any studies with human participants or animals performed by any of the authors. Informed consent was obtained from all individual participants included in the study.
- 7.Mehta, R., Egiazarian, K., Texture classification using dense micro-block difference (DMD). Proc. Asian Conf. Comput. Vis. 1643–1658, 2014.Google Scholar
- 8.Perronnin, F., Sánchez, J., and Mensink, T., Improving the fisher kernel for large-scale image classification. Proc. Eur. Conf. Comput. Vis. 143–156, 2010.Google Scholar
- 11.Xu, Y., Yang, X., Ling, H., and Ji, H., A new texture descriptor using Mul-tifractal analysis in multi-orientation wavelet pyramid. Proc. IEEE Conf. Comput. Vis. Pattern Recog. 161–168, 2010.Google Scholar
- 17.Bo, Z., Yu-Xin, J., and Qing, D., et al., Prospective observation of contrast-enhanced patterns of thyroid nodules with SonoVue[J]. Chinese Journal of Medical Imaging Technology, 2010.Google Scholar
- 19.Heng, Z, and Zheng, S., Research Progress of texture analysis in thyroid nodule imaging[J]. Chinese computed Medical Imaging, 2018.Google Scholar
- 20.Xia, Kai-Jian; Yin, Hong-Sheng; Rong, Gong-Sheng; Wang, Jiang-Qiang; Jin, Yong. X-ray image enhancement base on the improved adaptive low-pass filtering. J. Med. Imaging Health Inform., Volume 8, Number 7, September 2018, pp. 1342–1348(7), doi: https://doi.org/10.1166/jmihi.2018.2472 CrossRefGoogle Scholar
- 21.Cimpoi, M., Maji, S., Kokkinos, I., Mohamed, S., and Vedaldi, A., Describing textures in the wild. Proc. IEEE Conf. Comput. Vis. Pattern Recog. 3606–3613, 2014.Google Scholar
- 29.Sharma, G., ul Hussain, S., and Jurie, F., Local higher-order statistics (LHS) for texture categorization and facial analysis. Proc. Eur. Conf. Comput. Vis. 1–12, 2012.Google Scholar