AUTOMATIC CONSTRUCTION OF ABNORMAL SIGNAL EXTRACTION PROCESSING FROM 3D DIFFUSION WEIGHTED IMAGE
|
Accession number;07A0187200
|
| Title;AUTOMATIC CONSTRUCTION OF ABNORMAL SIGNAL EXTRACTION PROCESSING FROM 3D DIFFUSION WEIGHTED IMAGE |
| Author;
NAKANO YUTA
(Yokohama National Univ., Yokohama, Jpn)
NAGAO TOMOHARU
(Yokohama National Univ., Yokohama, Jpn)
TAKEBAYASHI SHIGEO
(Yokohama City Univ. Medical Center, Yokohama, Jpn)
|
Journal Title;IEIC Technical Report (Institute of Electronics, Information and Communication Engineers)
|
Journal Code:S0532B
|
ISSN:0913-5685
|
|
VOL.106;NO.448(IE2006 138-184);PAGE.141-146(2007)
|
| Figure&Table&Reference; |
| Pub. Country;Japan |
| Language;English |
| Abstract;In the field of image processing, we often deal with multi-dimensional data recently. Especially, we are required to process 3D image data, for instance CT, MRI and PET in medical image processing. It is generally difficult to construct 3D image processing compared with 2D ones. Thus, we previously proposed a method named "3D-ACTIT; Three Dimensional-Automatic Construction of Tree-structural Image Transformation" to make various 3D image processing procedures automatically. This system constructs tree-structured image processing filter with one-input one-output filter and two-inputs one-output filter by using Genetic Programming to satisfy the given several image examples. We currently research on 3D-DWI (Diffusion Weighted Image) data and its processing. DWI is a kind of MRI, visualizes the diffusion phenomenon of H2O molecules by strength of signal and given as 3D image data. This technology is expected as more effective method to detect cerebral infection and other diseases. It is easy to detect abnormal signals visually, however, high signal is not always abnormal signal in DWI. When we think about automatic diagnosis system, it is necessary to judge whether each high signal is abnormal or not in consideration of medical knowledge. In this paper, we apply 3D-ACTIT to 3D-DWI data and construct the processing which extracts only abnormal signal regions. To use medical knowledge in processing, we added image features specializing in DWI as 3D image processing filters to 3D-ACTIT system. (author abst.) |
|
|
|
Related Articles;
|
|
