講演名 2007-01-26
A comparison study of feature selection algorithms for automatic differentiation of obstructive lung diseases
,
PDFダウンロードページ PDFダウンロードページへ
抄録(和)
抄録(英) The study was conducted to test and compare the performance of several feature selection methods for the differentiation among obstructive lung diseases based on the features from texture and shape analysis using HRCT (High Resolution Computerized Tomography) images. Five different types of feature selection methods were tested and compared based on the classification performance and clinical applicability. Based on the experimental results, we investigate characteristics of those feature selection methods and evaluate in terms of classification accuracy, robustness, number of iteration and computation time. Especially, SFFS give users good advantage that can terminate the algorithm earlier without accuracy lost.
キーワード(和)
キーワード(英) Bayesian classifier / Feature Selection / Obstructive Lung Disease / Support vector machine
資料番号 MI2006-104
発行日

研究会情報
研究会 MI
開催期間 2007/1/19(から1日開催)
開催地(和)
開催地(英)
テーマ(和)
テーマ(英)
委員長氏名(和)
委員長氏名(英)
副委員長氏名(和)
副委員長氏名(英)
幹事氏名(和)
幹事氏名(英)
幹事補佐氏名(和)
幹事補佐氏名(英)

講演論文情報詳細
申込み研究会 Medical Imaging (MI)
本文の言語 ENG
タイトル(和)
サブタイトル(和)
タイトル(英) A comparison study of feature selection algorithms for automatic differentiation of obstructive lung diseases
サブタイトル(和)
キーワード(1)(和/英) / Bayesian classifier
第 1 著者 氏名(和/英) / Yu-Sub Sung
第 1 著者 所属(和/英)
Department of Industrial Engineering, Seoul National University College of Engineering
発表年月日 2007-01-26
資料番号 MI2006-104
巻番号(vol) vol.106
号番号(no) 509
ページ範囲 pp.-
ページ数 4
発行日