Respiratory motion correction method based on principal components analysis model for liver contrast-enhanced ultrasound sequences
10.7687/j.issn1003-8868.2017.09.005
- VernacularTitle:基于主成分分析模型的肝超声造影图像呼吸运动校正研究
- Author:
jie Jun ZENG
1
;
Ji ZHANG
;
rong Yan ZHANG
;
Juan CHEN
;
Feng XIAO
Author Information
1. 武汉大学中南医院医学影像科
- Keywords:
contrast-enhanced ultrasound;
principal component analysis;
respiratory motion;
VX2 tumor;
liver
- From:
Chinese Medical Equipment Journal
2017;38(9):5-11
- CountryChina
- Language:Chinese
-
Abstract:
Objective To relieve the influences of the respiratory motion on the liver contrast-enhance ultrasound (CEUS) image sequences,to enhance the quantitative analysis accuracy for liver CEUS and to put forward a correction strategy for the respiratory motion in liver CEUS sequences.Methods A principal component analysis (PCA) model of the respiratory motion in liver CEUS sequences was established with 18 cases of rabbit liver VX2 tumors,and a respiratory motion curve was generated based on the principal component with large data proportion,then the images with similar phases to the reference image were analyzed.Resnlts Correction made the mean structural similarity and mean correlation coefficient enhanced significantly to 0.57±0.11 and 0.78±0.11 respectively (P＜0.001),while the average of deviation valve (DV) was decreased to 29.9±7.02 which only was one-third of the original value.Threshold setting could further improve the quality of the selected image sequence.Conchusion The proposed respiratory motion method proves its effectiveness for rabbit liver CEUS image sequences,and thus contributes to enhancing the differential diagnosis rate of benign and malignant liver tumors.
