Evaluation studies of ultrasonic image features extraction for cervical lymph nodes.
- Author:
Junhua ZHANG
1
;
Yuanyuan WANG
;
Yi DONG
;
Yi WANG
;
Xinling SHI
Author Information
1. Dept. of Elec. Engn., Fudan University, Shanghai 200433, China.
- Publication Type:Journal Article
- MeSH:
Adult;
Aged;
Aged, 80 and over;
Evaluation Studies as Topic;
Female;
Humans;
Image Enhancement;
methods;
Image Interpretation, Computer-Assisted;
methods;
Image Processing, Computer-Assisted;
methods;
Lymph Nodes;
blood supply;
diagnostic imaging;
Male;
Middle Aged;
Neck;
Pattern Recognition, Automated;
Ultrasonography, Doppler;
Young Adult
- From:
Journal of Biomedical Engineering
2008;25(1):172-181
- CountryChina
- Language:Chinese
-
Abstract:
Abstract Ultrasonic image features of cervical lymph nodes were extracted by a computerized scheme. The scheme quantified the parameters of shape, border, cortex to medulla ratio, distributing of medulla, long axis to short axis ratio, vascular density and vascular pattern features. Then correlations between the experts andthe computerized scheme for assessing parameters in corresponding categories were computed to evaluate the validity of the parameters extracted by the computer. 106 cervical lymph nodes were studied, and the results showed that good agreement was seen in NRL (normalized radial length) zero crossing count, convex hull depth, long axis to short axis ratio, the vascular density and the entropy of the blood mapping boundary points' distribution.
