자기공명영상에서 신경회로망을 이용한 뇌경색의 시기 분류기 구현

박병래*1), 김화곤2)

1)부산가톨릭대학교 방사선학과
2)부산가톨릭대학교 방사선학과

Abstract : In this study, we conducted on neural network classifier for the steps of the cerebral infarction by means of objective-centered hierachial planning. A neural network classifier for the steps of cerebral infarction designed specially feature descriptors to distinguish hyperacute, acute, subacute in MR images. And using error back-propagation method for objective interpretation of the cerebral infarction steps. We expect thar this study performed as well as human experts and could br useful as clinical decision support tool for brain infarction patients. DWI(diffusion weighted image) is very exact for early diagnosis. We assumed period of infarction occurrence period condition of a disease to compare to analysis between region of disease and normal region in DWI, T2WI, FLAIR images.
keyword : Brain infarction, Diffusion, Magnetic Resonance Image

연제분류


첨부한 파일

Neural Network_RAE.hwp

첨부한 e-poster파일

투고내용을 삭제하시려면 "삭제"버튼을 누르시고 다시 접수하여 주시기 바랍니다.

 암 호

대한의료정보학회
Copyright (c) 2002 by The Korean Society of Medical Informatics