Presentation 2004/9/10
Preliminary Study on Support Vector Machine (SVM) and application to a detection algorithm of abnormal shadow candidates
Yoriko INENAGA, Satoshi KASAI, Koh MATSUI, Akiko KANO, Takeshi HARA, Hiroshi FUJITA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) We have been developing a Computer-Aided Diagnosis (CAD) system. Various features are utilized to distinguish abnormalities from normal tissues. Support Vector Machine (SVM), which is one of multivariate analyses, is recently watched with keen interest as one of the best techniques to obtain an optimal discrimination boundary. In this research, we investigated discrimination boundaries determined by a selected kernel function and its parameters as a preliminary study. SVM was applied to our detection algorithm of mass shadows on mammograms as an application. The results showed that this technique is useful on the algorithm.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Image Processing / Support Vector Machine / Mammogram / Mass Shadows
Paper # MI2004-39
Date of Issue

Conference Information
Committee MI
Conference Date 2004/9/10(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 Medical Imaging (MI)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Preliminary Study on Support Vector Machine (SVM) and application to a detection algorithm of abnormal shadow candidates
Sub Title (in English)
Keyword(1) Image Processing
Keyword(2) Support Vector Machine
Keyword(3) Mammogram
Keyword(4) Mass Shadows
1st Author's Name Yoriko INENAGA
1st Author's Affiliation R&D Center, Konica Minolta Medical & Graphic, Inc.()
2nd Author's Name Satoshi KASAI
2nd Author's Affiliation R&D Center, Konica Minolta Medical & Graphic, Inc.
3rd Author's Name Koh MATSUI
3rd Author's Affiliation R&D Center, Konica Minolta Medical & Graphic, Inc.
4th Author's Name Akiko KANO
4th Author's Affiliation R&D Center, Konica Minolta Medical & Graphic, Inc.
5th Author's Name Takeshi HARA
5th Author's Affiliation R&D Center, Konica Minolta Medical & Graphic, Inc.
6th Author's Name Hiroshi FUJITA
6th Author's Affiliation Department of Intelligent Image Information, Faculty of Regeneration and Advanced Medical Science Graduate School of Medicine, Gifu University
Date 2004/9/10
Paper # MI2004-39
Volume (vol) vol.104
Number (no) 318
Page pp.pp.-
#Pages 6
Date of Issue