Journal of Computer Aided Diagnosis of Medical Images
Online ISSN : 1347-9245
ISSN-L : 1347-9245
A quantitative analysis of preprocessing operations for lung nodules detection from chest X-ray CT images
Suguru KAWAJIRIShinji MIZUNOHotaka TAKIZAWAShinji YAMAMOTOSatoshi UMEDATohru MATSUMOTOTakeshi IINUMAYukio TATENO
Author information
JOURNAL FREE ACCESS

2007 Volume 11 Issue 1 Pages 1-9

Details
Abstract
For improvement of nodule detection in CT, we have evaluated image preprocessing employed before our Com-puter-aided diagnosis (CAD) system for lung cancers. The preprocessing method consists of the median and tophat fil-ters. The median filter can reduce high frequency noises, and the tophat filter can remove low frequency changing in density. This paper describes an evaluation method by use of artificial nodules. Experimental results show the relation-ship between the parameters for the filters and the sizes and attenuation of detectable nodules.
Content from these authors
© 2007 by Japan Society of Computer Aided Diagnosis of Medical Images
feedback
Top