The following information was submitted:
Transactions: WSEAS TRANSACTIONS ON COMPUTERS
Transactions ID Number: 52-724
Full Name: Yan Zhang
Position: Professor
Age: ON
Sex: Male
Address: HIT Campus of Shenzhen University Town, Xili, Shenzhen, Guangdong
Country: CHINA
Tel:
Tel prefix:
Fax:
E-mail address: ianzh@hit.edu.cn
Other E-mails:
Title of the Paper: a parallel image reconstructor for fan beam ct scanner
Authors as they appear in the Paper: Guowei Xue, Yan Zhang, Xiuwang Li
Email addresses of all the authors: xueguowei@hit.edu.cn, ianzh@hit.edu.cn, liuxiuwang@anketech.com
Number of paper pages: 10
Abstract: The acceleration of CT image reconstruction is a key technology to satisfy the underlying real time demand for CT scanner. Nowadays most accelerating methods consider the image reconstructor as a separate subsystem. But with the development of CT technology, the amount of projection data is becoming larger and larger which makes the transferring of projection data takes much more time. In this paper, we propose a new parallel reconstruction method for fan beam X-ray transmission CT considering the data transferring and data processing as a whole. The advantage of the method is that the intermediate CT image can be refreshed on the arriving of each single projection frame. The CT image is generated with the arrival of the final projection frame and the last refreshment of the intermediate image. The performance of the method is evaluated by reconstructing the water phantom on a fan beam CT scanner. The experimental results show that the reconstruction time(including!
the data transfer time) can be reduced from 5 seconds to approximately 4 seconds, saving up to 20% time. The spatial resolution and the low contrast resolution show that the image quality is acceptable in clinic.
Keywords: Fan beam CT; Image reconstruction; Data transfer; Parallel algorithm; Speed up
EXTENSION of the file: .pdf
Special (Invited) Session:
Organizer of the Session:
How Did you learn about congress: CT image reconstructin, parallel algorithm
IP ADDRESS: 183.37.139.226
