작성자 M.I.D
[논문] Computer-Aided Diagnosis for Liver Cirrhosis using Texture featur…
본문
Computer-Aided Diagnosis for Liver Cirrhosis using Texture features Information Analysis in Computed Tomography
(컴퓨터단층영상에서 TIA를 이용한 간경화의 컴퓨터보조진단)
The Journal of the Korea Contents Association (한국콘텐츠학회논문지)
Volume 12 Issue 4 / Pages.358-366 / 2012 / 1598-4877(pISSN) / 2508-6723(eISSN)
The Korea Contents Association (한국콘텐츠학회)
김창수 /Kim, Chang-Soo (부산가톨릭대학교 보건과학대학 방사선학과) ; 고성진 / Ko, Seong-Jin (부산가톨릭대학교 보건과학대학 방사선학과) ; 강세식 / Kang, Se-Sik (부산가톨릭대학교 보건과학대학 방사선학과) ; 김정훈 / Kim, Jung-Hoon (부산가톨릭대학교 보건과학대학 방사선학과) ; 김동현 / Kim, Dong-Hyun (부산가톨릭대학교 보건과학대학 방사선학과) ; 최석윤 / Choi, Seok-Yoon (부산가톨릭대학교 보건과학대학 방사선학과)
Cirrhosis is a consequence of chronic liver disease characterized by replacement of liver tissue by fibrosis, scar tissue and regenerative nodules leading to loss of liver function. Liver Cirrhosis is most commonly caused by alcoholism, hepatitis B and C, and fatty liver disease, but has many other possible causes. Some cases are idiopathic disease from unknown cause. Abdomen of liver Computed tomography(CT) is one of the primary imaging procedures for evaluating liver disease such as liver cirrhosis, Alcoholic liver disease(ALD), cancer, and interval changes because it is economical and easy to use. The purpose of this study is to detect technique for computer-aided diagnosis(CAD) to identify liver cirrhosis in abdomen CT. We experimented on the principal components analysis(PCA) algorithm in the other method and suggested texture information analysis(TIA). Forty clinical cases involving a total of 634 CT sectional images were used in this study. Liver cirrhosis was detected by PCA method(detection rate of 35%), and by TIA methods(detection rate of 100%-AGI, TM, MU, EN). Our present results show that our method can be regarded as a technique for CAD systems to detect liver cirrhosis in CT liver images.
(컴퓨터단층영상에서 TIA를 이용한 간경화의 컴퓨터보조진단)
The Journal of the Korea Contents Association (한국콘텐츠학회논문지)
Volume 12 Issue 4 / Pages.358-366 / 2012 / 1598-4877(pISSN) / 2508-6723(eISSN)
The Korea Contents Association (한국콘텐츠학회)
김창수 /Kim, Chang-Soo (부산가톨릭대학교 보건과학대학 방사선학과) ; 고성진 / Ko, Seong-Jin (부산가톨릭대학교 보건과학대학 방사선학과) ; 강세식 / Kang, Se-Sik (부산가톨릭대학교 보건과학대학 방사선학과) ; 김정훈 / Kim, Jung-Hoon (부산가톨릭대학교 보건과학대학 방사선학과) ; 김동현 / Kim, Dong-Hyun (부산가톨릭대학교 보건과학대학 방사선학과) ; 최석윤 / Choi, Seok-Yoon (부산가톨릭대학교 보건과학대학 방사선학과)
Cirrhosis is a consequence of chronic liver disease characterized by replacement of liver tissue by fibrosis, scar tissue and regenerative nodules leading to loss of liver function. Liver Cirrhosis is most commonly caused by alcoholism, hepatitis B and C, and fatty liver disease, but has many other possible causes. Some cases are idiopathic disease from unknown cause. Abdomen of liver Computed tomography(CT) is one of the primary imaging procedures for evaluating liver disease such as liver cirrhosis, Alcoholic liver disease(ALD), cancer, and interval changes because it is economical and easy to use. The purpose of this study is to detect technique for computer-aided diagnosis(CAD) to identify liver cirrhosis in abdomen CT. We experimented on the principal components analysis(PCA) algorithm in the other method and suggested texture information analysis(TIA). Forty clinical cases involving a total of 634 CT sectional images were used in this study. Liver cirrhosis was detected by PCA method(detection rate of 35%), and by TIA methods(detection rate of 100%-AGI, TM, MU, EN). Our present results show that our method can be regarded as a technique for CAD systems to detect liver cirrhosis in CT liver images.