電腦輔助系統應用於活體肝臟移受贈者肝臟影像分割評估

The Study of Automated Estimation of Recipient Liver Volume for Liver Transplantation

蔣獻文(Hsien-Wen Chiang) ； 鄭汝汾(Yu-Fan Cheng) ； 陳肇隆(Chao-Long Chen) ； 莊子肇(Tzu-Chao Chuang) ； 蔣獻仁(Hsien-Jen Chiang) ； 李榮輝(Jung-Hui Lee)

《中華放射線技術學雜誌》 37卷2期 (2013/06) Pp. 74-82

摘要

The purpose of this project is to establish an automated computer-assisted system to measure the recipient liver volume and comparing the disparity between automated CT volumetric measurements and the actual recipient liver volume. The estimation of liver volume uses CT volumetric measurements. However, the region of liver on CT imaging was chosen manually by experienced senior radiation technologist or radiologist. Therefore, different people will acquire different results of liver volume estimation by CT volumetric measurements not to mention the fact that it costs manpower and time-consuming. Automated measurement of liver volume by computer-assisted system will not only reduce the interobserver variation but also save the human resources. Income of 30 patients of the study comparing traced manually by a radiologist. T Automated measurement of liver volume by computer-assisted system using region growing algorithm will not only reduce the interobserver variation but also save the human resources.he liver volumes extracted by the computerized scheme were compared to those traced manually by a radiologist, used as ”gold standard.” region growing algorithm are compared with the recipient liver removed after surgery to do a comparison, the statistical aspects of the use of paired samples T-test and the Pearson correlation coefficient test. The results showed that Pearson correlation coefficients for computer-assisted system between manual and automatic segmentations were 0.986, p-value＜0.05 respectively.

關鍵字

肝臟分割 ； 區域生長演算法 ； 電腦輔助系統

並列摘要

The purpose of this project is to establish an automated computer-assisted system to measure the recipient liver volume and comparing the disparity between automated CT volumetric measurements and the actual recipient liver volume. The estimation of liver volume uses CT volumetric measurements. However, the region of liver on CT imaging was chosen manually by experienced senior radiation technologist or radiologist. Therefore, different people will acquire different results of liver volume estimation by CT volumetric measurements not to mention the fact that it costs manpower and time-consuming. Automated measurement of liver volume by computer-assisted system will not only reduce the interobserver variation but also save the human resources. Income of 30 patients of the study comparing traced manually by a radiologist. T Automated measurement of liver volume by computer-assisted system using region growing algorithm will not only reduce the interobserver variation but also save the human resources.he liver volumes extracted by the computerized scheme were compared to those traced manually by a radiologist, used as ”gold standard.” region growing algorithm are compared with the recipient liver removed after surgery to do a comparison, the statistical aspects of the use of paired samples T-test and the Pearson correlation coefficient test. The results showed that Pearson correlation coefficients for computer-assisted system between manual and automatic segmentations were 0.986, p-value＜0.05 respectively.

並列關鍵字

liver segmentation ； region growing algorithm ； computer-assisted system