DIAGNOSTIC CLASSIFICATION OF AUTOANTIBODY REPERTOIRES IN ENDOCRINE OPHTHALMOPATHY USING AN ARTIFICIAL NEURAL-NETWORK

Purpose: The aim of this study was to classify the human Ige autoantib ody repertoire of sera from patients suffering from endocrine ophthalm opathy (EOP) and healthy subjects (CTRL) for diagnostic purposes using the recently developed Megablot technique. This technique allows for the simultaneous and quantitative screening of a large set of antigens and uses multivariate statistical techniques and an artificial neural network. Methods: Sera were tested against Western blots (WBs) of SDS -PAGE preparations of proteins from human extraorbital eye muscle (EOP : n-16; CTRL: n=11). Digital image analysis was performed. The blots w ere subsequently analyzed by multivariate statistical techniques (anal ysis of discriminance) and an artificial neural network (probalistic n eural network). Results: The sera of both the EOP and CTRL groups show ed a complex staining pattern against WBs of SDS-PAGEs from human eye muscle. Using the multivariate statistical technique for classificatio n, all of the known samples and 85% of the unknown samples (not presen ted during calculation) were assigned to their correct clinical group. Using the artificial neural network as classifier, all of the samples presented during training and 96.3% of the unknown samples (not train ed) were assigned correctly. Conclusions: The artificial neural networ k exceeds the ability of multivariate statistical techniques such as a nalysis of discriminance to assign unknown samples to their correct pr edefined group. Thus, the neural network exceeds other methods in gene ralizing some similarities of blots used for classification. This stud y reveals that our new technique and irs evaluation using a neural net work call be used as a helpful diagnostic tool in autoimmune diseases such as endocrine ophthalmopathy.