DocumentCode :
2311083
Title :
A new automatic detection approach for hepatocellular carcinoma using C-acetate positron emission tomography
Author :
Chen, Sirong ; Wong, Longkin ; Feng, Dagan
Author_Institution :
Dept of Electron. & Inf. Eng., Hong Kong Polytech. Univ., China
Volume :
1
fYear :
2003
fDate :
14-17 Sept. 2003
Abstract :
Functional imaging techniques such as positron emission tomography (PET) has the potential for early diagnosis of malignant tumors. However, 40-50% of hepatocellular carcinoma (HCC), a common malignancy worldwide, can hardly be detected by the widely used F-2-fluoro-2-deoxy-D-glucose (FDG) PET. C-acetate PET has recently been found effective for detecting HCC. To perform quantitative analysis to obtain the diagnosis information, regions of interest (ROls) are needed to be extracted. Manual placement of ROIs is subject to operator´s skill and time-consuming. Furthermore, the small sizes of some ROIs make the task even more difficult. In this paper, we propose an approach to segment the dynamic C-acetate PET liver images automatically. The curves extracted from some segmented ROIs are then fitted to the presented C-acetate liver model. Finally, the parameter K, which has been validated as an indicator for detecting HCC, can be calculated.
Keywords :
image segmentation; liver; medical image processing; pattern clustering; positron emission tomography; tumours; automatic image segmentation; cluster analysis; dynamic C-acetate PET liver image; functional imaging technique; hepatocellular carcinoma; malignant tumor diagnosis; positron emission tomography; regions of interest; Biochemistry; Biomedical signal processing; Blood; Cancer; Computed tomography; Image segmentation; Liver neoplasms; Malignant tumors; Pixel; Positron emission tomography;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Image Processing, 2003. ICIP 2003. Proceedings. 2003 International Conference on
ISSN :
1522-4880
Print_ISBN :
0-7803-7750-8
Type :
conf
DOI :
10.1109/ICIP.2003.1247150
Filename :
1247150
Link To Document :
بازگشت