A New Digital Image Processing Technique for Mixed Pattern Analysis of Anti-neutrophil Cytoplasmic Antibody Test.
- Author:
Jung UK SIR
1
;
Think You KIM
Author Information
1. Department of Diagnostic Immunology / Laboratory Medicine, The Hospital for Rheumatic Diseases, Hanyang University Medical Center, Seoul, Korea. tykim@hanyang.ac.kr
- Publication Type:Original Article
- Keywords:
Computer assisted image processing;
Anti-neutrophil cytoplasmic antibody;
Anti-DNA antibodies;
Antinuclear antibody;
Indirect immunofluorescence assay
- MeSH:
Antibodies;
Antibodies, Antineutrophil Cytoplasmic*;
Antibodies, Antinuclear;
Autoantibodies;
Autoimmune Diseases;
Classification;
Diagnosis;
Diagnosis, Differential;
Fluorescent Antibody Technique, Indirect;
Image Processing, Computer-Assisted;
Neutrophils;
Rheumatic Diseases;
Systemic Vasculitis
- From:The Journal of the Korean Rheumatism Association
2004;11(4):333-341
- CountryRepublic of Korea
- Language:Korean
-
Abstract:
OBJECTIVE: Anti-neutrophil cytoplasmic antibody (ANCA) is an important marker for the diagnosis and the classification of rheumatic diseases and systemic vasculitis. If the autoantibodies that can be stained in the nucleus of neutrophil used for substrate of ANCA test, there's a need for its differential diagnosis from real p-ANCA. This paper focuses on digital image processing technique for the differentiation of p-ANCA without performing additional tests. METHODS: Positive ANCA results which showed mixed fluorescent pattern were transformed into digital image. Using Matlab (MathWorks, U.S.A.), we developed 2D to 3D transformation method and virtual tomography for the interpretation of mixed fluorescent pattern, and compared these results with ANA and anti-DNA test by indirect immunofluorescence (IIF) method using IT-AIT, IT-ANCA, and IT-DNA kit (ImmunoThink, Korea). RESULTS: By applying the 3D transformation method and virtual tomography to the results of ANCA test where combined antibodies exist, we were able to separate each different fluorescent pattern that were difficult to separate by manual reading. CONCLUSION: The new digital image analysis methods developed in this study displace some of the disadvantages of IIF method. Therefore, these methods can easily be applied to complex samples, and can allow rapid and accurate tests for rheumatic diseases and other autoimmune diseases.
