Thyroid Image Feature Extraction and State Recognition

被引：0

作者：

Wang, Mei ^{[1
]}

Wu, Xiaowei ^{[1
]}

Xu, Jian ^{[1
]}

机构：

[1] Xian Univ Sci & Technol, Coll Elect & Control Engn, Xian 710054, Peoples R China

来源：

INTELLIGENT SCIENCE AND INTELLIGENT DATA ENGINEERING, ISCIDE 2011 | 2012年 / 7202卷

关键词：

Feature extraction; Thyroid; K-nearest neighbors; Recognition; Medicine image;

D O I：

暂无

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Aiming at the medical monitoring problem, the thyroid image features are extracted in this paper and the human thyroid CT image is classified. Firstly, the area of the two thyroid leaves and the cross section of thyroid biopsy are used as original features. Secondly, the area ratio of the two thyroid leaves and the cross section area of the two thyroid leaves are calculated and serve as the new features which describes the pattern of human thyroid. After the feature extraction, the K-nearest neighbor classification algorithm is adopted to distinguish images belonging to the normal or the abnormal categories respectively. The experimental result shows that the feature extraction method is suitable and the recognition result is generated with a certain accuracy, reliability and practicability when these two features to classify the normal and abnormal categories.

引用

页码：273 / 277

页数：5