Presentation 2005-01-21
Improvements of a Breast Cancer Detection System for Three Dimensional Breast CT Images by Using New Feature and Classification Method
Akihiro Hamada, Akinobu Shimizu, Hidefumi Kobatake, Kunihisa Miyakawa, Natsuko Sato,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper describes improvements of breast cancer detection system for three dimensional (3D) breast CT images. To improve the performance of classification in this system, we have increased number of features from 237 to 520 and developed a new classifier using Support Vector Machine so as to improve discrimination accuracy. In order to evaluate performance of the improved system, we applied the system to thirty five multi-slice CT images (including 34 cancers) and achieved 94.1% sensitivity with an average of 0.06 false positive detection per image, which is 5.9% sensitivity improvement and 0.31 false positive reduction.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Computer-Aided Diagnosis / Three Dimensional CT image / Breast Cancer / Cancer Classification
Paper # MI2004-59
Date of Issue

Conference Information
Committee MI
Conference Date 2005/1/14(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) Improvements of a Breast Cancer Detection System for Three Dimensional Breast CT Images by Using New Feature and Classification Method
Sub Title (in English)
Keyword(1) Computer-Aided Diagnosis
Keyword(2) Three Dimensional CT image
Keyword(3) Breast Cancer
Keyword(4) Cancer Classification
1st Author's Name Akihiro Hamada
1st Author's Affiliation Graduate School of Bio-Applications and Systems Engineering, Tokyo University of Agriculture and Technology()
2nd Author's Name Akinobu Shimizu
2nd Author's Affiliation Graduate School of Bio-Applications and Systems Engineering, Tokyo University of Agriculture and Technology
3rd Author's Name Hidefumi Kobatake
3rd Author's Affiliation Graduate School of Bio-Applications and Systems Engineering, Tokyo University of Agriculture and Technology
4th Author's Name Kunihisa Miyakawa
4th Author's Affiliation Department of Radiology, National Cancer Center Hospital
5th Author's Name Natsuko Sato
5th Author's Affiliation Department of Radiology, National Cancer Center Hospital
Date 2005-01-21
Paper # MI2004-59
Volume (vol) vol.104
Number (no) 579
Page pp.pp.-
#Pages 6
Date of Issue