Automated Retrieval of CT Images of Liver Lesions on the Basis of Image Similarity: Method and Preliminary Results1

Content-based retrieval of images containing similar-appearing lesions is practical; preliminary assessment of this approach shows reasonable retrieval results compared with an independently constructed pairwise visual similarity standard for three types of liver lesions visible on portal venous CT...

Full description

Saved in:
Bibliographic Details
Published inRadiology Vol. 256; no. 1; pp. 243 - 252
Main Authors Napel, Sandy A., Beaulieu, Christopher F., Rodriguez, Cesar, Cui, Jingyu, Xu, Jiajing, Gupta, Ankit, Korenblum, Daniel, Greenspan, Hayit, Ma, Yongjun, Rubin, Daniel L.
Format Journal Article
LanguageEnglish
Published Radiological Society of North America, Inc 01.07.2010
Subjects
Original Research
Online AccessGet full text

Cover

More Information
Summary:Content-based retrieval of images containing similar-appearing lesions is practical; preliminary assessment of this approach shows reasonable retrieval results compared with an independently constructed pairwise visual similarity standard for three types of liver lesions visible on portal venous CT images.
Bibliography:Author contributions: Guarantors of integrity of entire study, S.A.N., C.F.B., H.G., D.L.R.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, S.A.N., C.F.B., H.G., Y.M.; clinical studies, C.F.B., H.G.; statistical analysis, S.A.N., J.C., J.X., H.G., Y.M.; and manuscript editing, S.A.N., C.F.B., J.C., J.X., H.G., D.L.R.
ISSN:0033-8419
1527-1315
DOI:10.1148/radiol.10091694