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

Abstract

Background: This study explored the diagnostic potential of image-processing analysis in optical coherence tomography (OCT) images to detect systemic vascular changes in individuals with systemic diseases.

Methods: Ocular OCT images from two cohorts diabetic patients and healthy control subjects were analyzed. A novel Superpixel Segmentation (SpS) algorithm was used to process these images and extract optical image density information from ocular vascular tissue. The algorithm was applied to isolate the choroid layer for analysis of its optical properties. The procedure was performed by separate examiners, and both inter- and intra-observer repeatability were assessed. Choroidal area (CA) and choroidal optical image density (COID) metrics were used to assess structural changes in the vascular tissue and predict alterations in the choroidal parameters.

Results: A total of 110 diabetic patient eye images and 92 healthy control images were processed. The results showed significant differences in CA and COID between diabetic and healthy eyes, indicating that these parameters could serve as valuable biomarkers for early vascular damage.

Conclusions: The use of the SpS algorithm on OCT B-scan images allows for the identification of new parameters linked to ocular vascular damage. These findings suggest that digital image-processing techniques can reveal differences in vascular tissue, offering potential new indicators of pathology.

Keywords: OCT; algorithm; diabetic; optical density.

Figure 1

Image outputs obtained via Superpixel algorithm processing for one OCT B-scan image. The upper image (a) shows segmentation, and the lower image (b) shows the choroid tissue boundaries.

Figure 2

Image outputs obtained via Superpixel algorithm processing for one OCT B-scan image. The upper pictures are from healthy subjects, and the lower pictures are from diabetic patients. The left column (a) shows OCT B-scan images for both types of subject, the center column (b) shows segments with the same semantic information and the right column (c) shows choroid delimited tissue.