Presentation 2004/5/21
Advanced discrimination system of tumors on mammograms based on bilateral subtraction features (Second report)
Akinobu SHIMIZU, Daisuke HIRAI, Mitsutaka NEMOTO, Hidefumi KOBATAKE, Shigeru NAWANO,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In the previous report, we have developed a process based on comparison of bilateral mammograms and confirmed its effectiveness. In this process, 93 kinds of features were calculated from the bilateral mammograms and subtraction values between the features were computed as proposed features, called subtraction features. A sub-optimal feature set was selected from 186 features which consisted of 93 original features and 93subtraction features in order to discriminate candidates corresponding to tumors from those corresponding to normal tissues. This paper describes results of increasing number of features to 955 by adding four kinds of images, such as differential images of original mammograms. In the discrimination experiment using 1,008 mammograms, it was confirmed that the number of false positives was less than one third of the previous number.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) computer-aided diagnosis system / comparison of bilateral breast / mammogram / tumor shadow
Paper # PRMU2004-25,MI2004-25,WIT2004-25
Date of Issue

Conference Information
Committee WIT
Conference Date 2004/5/21(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Well-being Information Technology(WIT)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Advanced discrimination system of tumors on mammograms based on bilateral subtraction features (Second report)
Sub Title (in English)
Keyword(1) computer-aided diagnosis system
Keyword(2) comparison of bilateral breast
Keyword(3) mammogram
Keyword(4) tumor shadow
1st Author's Name Akinobu SHIMIZU
1st Author's Affiliation Institute of Symbiotic Science and Technology, Tokyo University of Agriculture and Technology()
2nd Author's Name Daisuke HIRAI
2nd Author's Affiliation Institute of Symbiotic Science and Technology, Tokyo University of Agriculture and Technology
3rd Author's Name Mitsutaka NEMOTO
3rd Author's Affiliation Institute of Symbiotic Science and Technology, Tokyo University of Agriculture and Technology
4th Author's Name Hidefumi KOBATAKE
4th Author's Affiliation Institute of Symbiotic Science and Technology, Tokyo University of Agriculture and Technology
5th Author's Name Shigeru NAWANO
5th Author's Affiliation National Cancer Center Hospital East
Date 2004/5/21
Paper # PRMU2004-25,MI2004-25,WIT2004-25
Volume (vol) vol.104
Number (no) 93
Page pp.pp.-
#Pages 6
Date of Issue