Title
A variational approach for reconstructing low dose images in clinical helical CT
Abstract
Exposure to medical radiation is an important healthcare concern. In this paper we present a variational formulation for image reconstruction from low-dose transmission X-ray CT as well as its application to real clinical data. Our projection domain reconstruction method is based on high-order total variation penalties. We apply the method to clinical neck data obtained from a current 64-channel MDCT helical scanner, where small features and large dynamic range variation are pervasive. The ability to generate high-quality images even at 75% dose reduction is demonstrated. In particular, our approach exhibits image SNR comparable to full-dose filtered backprojection images while avoiding the obvious stair case and blockiness artifacts typically encountered in such methods.
Year
DOI
Venue
2010
10.1109/ISBI.2010.5490057
ISBI
Keywords
Field
DocType
neck,cone-beam spiral,high-order total variation penalty,computerised tomography,low dose image,64-channel mdct helical scanner,filtered backprojection image,large dynamic range variation,variational approach,image snr,full-dose filtered backprojection images,mdct,high-quality image,high-order total variation penalties,real clinical data,blockiness artifacts,image reconstruction,stair case artifacts,low dose imaging,low-dose transmission x-ray ct,iterative reconstruction technique,clinical helical,projection domain reconstruction method,clinical neck data,medical image processing,total variation,noise,iterative methods,computed tomography,biomedical imaging,image restoration,tv,iterative reconstruction,dynamic range
Iterative reconstruction,Computer vision,Image generation,Pattern recognition,Iterative method,Medical imaging,Computer science,Computed tomography,Artificial intelligence,Scanner,Image restoration,Large dynamic range
Conference
ISSN
ISBN
Citations 
1945-7928 E-ISBN : 978-1-4244-4126-6
978-1-4244-4126-6
0
PageRank 
References 
Authors
0.34
3
5
Name
Order
Citations
PageRank
Synho Do19412.86
W. Clem Karl221.74
Mannudeep Kalra314414.28
Thomas J. Brady432.44
Homer Pien511310.40