Title | ||
---|---|---|
Texture analysis and classification of diffuse thyroid diseases based on ultrasound images |
Abstract | ||
---|---|---|
This paper describes a method of classification of diffuse thyroid disease based on ultrasound images. Our method proposes a new Co-occurrence Matrix named as Wavelet Multi-sub-bands Co-occurrence Matrix (WMCM), based on which a series of new texture features can be calculated. A new feature of fibrous variant texture is also proposed. The mRMR method is used to select an effective feature set from the feature space composed by those feature we proposed and other common texture feature. 180 samples of thyroid ultrasound images have been classified based on the selected feature set. Classification accuracies of Normal, Graves disease and Hashimotou0027s disease with SVM classifier are 87.83%, 83.67%, and 92.17%, respectively. It can be seen from contrast experiments that the features proposed in this paper can improve the classification accuracy by 5% or more comparing to other features. |
Year | DOI | Venue |
---|---|---|
2018 | 10.1109/i2mtc.2018.8409650 | instrumentation and measurement technology conference |
Field | DocType | Citations |
Feature vector,Pattern recognition,Image texture,Thyroid,Electronic engineering,Feature extraction,Feature set,Artificial intelligence,Medicine,Ultrasound,Thyroid disease,Wavelet | Conference | 0 |
PageRank | References | Authors |
0.34 | 3 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Dandan Li | 1 | 27 | 7.99 |
Zhang Yakui | 2 | 0 | 0.34 |
Du Linyao | 3 | 0 | 0.34 |
Zhou Xianli | 4 | 0 | 0.34 |
shen yi | 5 | 0 | 0.68 |