Computational image analysis for prognosis determination in DME. (October 2017)

• Record Type: Journal Article
• Authors: Gerendas, Bianca S.; Bogunovic, Hrvoje; Sadeghipour, Amir; Schlegl, Thomas; Langs, Georg; Waldstein, Sebastian M.; Schmidt-Erfurth, Ursula
• Is Part Of: Vision research. Volume 139(2017)
• Page Start: 204
• Page End: 210
• Publication Date: 2017-10
• Subjects: Machine learning -- Large-scale data analysis -- Diabetic macular edema -- Random forest -- Prediction -- Computational image analysis; Vision -- Periodicals; 573.88
• DOI: 10.1016/j.visres.2017.03.008
• Languages: English
• ISSNs: 0042-6989

Abstract

Abstract: In this pilot study, we evaluated the potential of computational image analysis of optical coherence tomography (OCT) data to determine the prognosis of patients with diabetic macular edema (DME). Spectral-domain OCT scans with fully automated retinal layer segmentation and segmentation of intraretinal cystoid fluid (IRC) and subretinal fluid of 629 patients receiving anti-vascular endothelial growth factor therapy for DME in a randomized prospective clinical trial were analyzed. The results were used to define 312 potentially predictive features at three timepoints (baseline, weeks 12 and 24) for best-corrected visual acuity (BCVA) at baseline and after one year used in a random forest prediction path. Preliminarily, IRC in the outer nuclear layer in the 3-mm area around the fovea seemed to have the greatest predictive value for BCVA at baseline, and IRC and the total retinal thickness in the 3-mm area at weeks 12 and 24 for BCVA after one year. The overall model accuracy was R 2 = 0.21/0.23 (p < 0.001). The outcomes of this pilot analysis highlight the great potential of the proposed machine-learning approach for large-scale image data analysis in DME and other retinal diseases.