Macular Telangiectasia Type 2

• DOI: 10.1016/j.xops.2022.100261
• Publication Date: 2022-12-09T01:07:52Z
• Authors: Emily Y. Chew; Tunde Peto; Traci E. Clemons; Ferenc B. Sallo; Daniel Pauleikhoff; Irene Leung; Glenn J. Jaffe; Tjebo F.C. Heeren; Catherine A. Egan; Peter Charbel Issa; Konstantinos Bala...; Frank G. Holz; Alain Gaudric; Alan C. Bird; Martin Friedlander
• Keywords: Fundus (uterus); Macular telangiectasia; Autofluorescence

To develop a severity classification for macular telangiectasia type 2 (MacTel) disease using multimodal imaging.An algorithm was used on data from prospective natural history study of MacTel development.A total 1733 participants enrolled in an international MacTel.The Classification and Regression Trees (CART), predictive nonparametric machine learning, analyzed the features imaging important development classification, including reading center gradings following digital images: stereoscopic color red-free fundus photographs, fluorescein angiographic images, autofluorescence spectral-domain (SD)-OCT images. models that least square method created decision tree ocular images into different categories severity.The primary target interest by CART change best-corrected visual acuity (BCVA) at baseline right left eyes. These analyses were repeated BCVA obtained last visit eyes.The demonstrated 3 classification: OCT hyper-reflectivity, pigment, ellipsoid zone loss. By combining these (as absent, present, noncentral involvement, central involvement macula), 7-step scale created, ranging excellent to poor acuity. At grade 0, are not present. most severe grade, pigment exudative neovascularization further validate Generalized Estimating Equation regression models, annual relative risk progression over period 5 years vision loss along performed.This analysis current modalities followed informed featuring variables SD-OCT. This is designed provide better communications other clinicians, researchers, patients.Proprietary or commercial disclosure may be found after references.