167 “Journal of the Korean Society of Radiology, Volume 6, Number 3” Feature Extraction of Disease Region in Stomach Images Based on DCT Byeoungju Ahn, SangBock Lee Dept. of Radiology, Nambu University DCT기반 위장영상 질환부위의 특징추출 안병주, 이상복 남부대학교 방사선학과 Abstract In this paper, we present an algorithm to extract features about disease region in digital stomach images. For feature extraction, DCT coefficients of gastrointestinal imaging matrix was obtained. DCT coefficent matrix is concentrated energy in low frequency region, we were extracted 128 feature parameters in low frequency region. Extracted feature parameters can using for differential compression of PACS and, can using for input parameter in CAD. Key Words : Digital Stomach Image, Image plate, Feature Extraction, DCT 요약 본 논문에서는 의용영상의 병소부위 특징을 추출하는 알고리즘을 제시하였다. 특징 추출을 위해 위장영상을 입력하 여 DCT계수 행렬을 구하였다. DCT계수 행렬은 저주파 영역으로 에너지가 집중되기 때문에 저주파 영역에서 128개의 특징 파라미터를 추출하였다. 추출된 특징 파라미터를 이용하여 질환영상과 정상영상을 비교하여 그래프로 나타내었 다. 특징 파라미터는 PACS의 차등압축과 CAD를 위한 입력 파라미터로 활용될 수 있을 것이다. 중심단어: 디지털위장영상, 이미지 플레이트, 특징추출, DCT I. Introduction Now, a study about image and multimedia information system is actively progressed. Using these results of a study, many researchers are studying for applying to the medical fields. Many researcher is concerning about PACS which is able to process mass digital medical image data [1] . Especially, success of modern hospitals depends on excellent development and management of PACS. IF PACS is popularized and ultrahigh speed communication line is equipped, all patients can be supported to excellent doctors using the PACS and ultrahigh speed communication line. And it is able to improve the level of medical diagnosis. If doctor use sample DB(Database) medical image, discrimination about medical image is automatically made possible. So in this paper we present an algorithm to extract features about disease region in digital stomach image. The results of study will be applied Corresponding Author: Sang Bock Lee E-mail: sblee@nambu.ac.kr, Tel: +82-62-970-0156 Addr. Dept. of Radiology, Nambu University, 23 Chumdan Jungang-ro, Gwangsan-gu, Gwangju, Korea Received : March 25, 2012 Revised : June 07, 2012 Accepted : June 22, 201