A Superpixel-Based Algorithm for Detecting Optical Density Changes in Choroidal OCT Images of Diabetic Patients

Purpose: This study explores the diagnostic potential of applying image-processing analysis to optical coherence tomography (OCT) images to detect systemic vascular changes in individuals with systemic diseases. Methods: The study enrolled analyzed ocular OCT images obtained from two cohorts: diabetic patients and healthy control subjects. The novel Superpixel Segmentation (SpS) algorithm developed for this paper was used to process these images and extract optical image density information from ocular vascular tissue. This procedure was employed in the image segmentation phase to isolate the choroid layer for subsequent analysis of the images’ optical properties. Separate examiners performed the image treatment and processing procedure and both inter- and intra-observer method repeatability were assessed. Choroidal area (CA) and choroidal optical image density (COID) metrics were used to assess structural alterations in the vascular tissue and predict changes to analytical choroidal parameters. Results: A total of 110 diabetic patient eye images and 92 healthy subjects eye images were processed. The results revealed significant differences in CA and COID between diabetic and healthy eyes. In short, detected image density remained unaffected by the extent of the choroidal tissue analyzed, indicating that these parameters could serve as valuable biomarkers of early vascular damage in OCT images. Conclusions: Use of the SpS algorithm on OCT B-scan images allows for the determination of a new parameter linked to ocular vascular damage. Translational relevance: The application of digital image processing techniques reveals differences in vascular tissue in OCT images, offering potential new indicators of pathology.