advertisement
Abstract #5227 Published in IGR 1-2
(Influence of myopic disc shape in a classification program of the Heidelberg retina tomograph)
Yamazaki Y; Yoshikawa K; Kunimatsu S; Koseki N; Suzuki Y; Matsumoto S; Araie MNippon Ganka Gakkai Zasshi 1999; 103: 392-398
PURPOSE: We investigated the influence of myopic disc shape on the diagnostic capability of a glaucoma diagnostic software (classification program) of the Heidelberg retina tomograph (HRT). Subjects: 66 eyes of 66 normal subjects and 78 eyes of 78 patients with open-angle glaucoma were studied. The criterion of glaucoma was a visual field defect appearing between Aulhorn classification stage II and stage V regardless of the maximum intraocular pressure value. METHODS: The subjects were divided into eyes with a myopic disc and those with a non-myopic disc on the basis of stereo disc fundus photographs without considering the refractive errors. Agreement between the classification program and the clinical diagnosis was evaluated by the calculation of sensitivity, specificity, and diagnostic precision, and the influence of the disc shape on HRT disc shape parameters was also evaluated. RESULTS: Sensitivity, specificity, and diagnostic precision of the classification program were 83%, 95%, and 89% in the eyes with a non-myopic disc, and 71%, 96%, and 83% in the eyes with a myopic disc, respectively. Rim volume, height variation contour, mean RNFL (retinal nerve fiber layer) thickness, and RNFL cross section area were significantly larger in the eyes with a myopic disc than in those with a non-myopic disc regardless of the clinical diagnosis. CONCLUSION: The classification program should be modified to adjust to a myopia-like disc shape in order to improve the capability of the glaucoma predictive diagnosis.LA: Japanese
Department of Ophthalmology, Nihon University School of Medicine, Tokyo, Japan.
Classification:
2.14 Optic disc (Part of: 2 Anatomical structures in glaucoma)
6.9.1 Laser scanning (Part of: 6 Clinical examination methods > 6.9 Computerized image analysis)
