Posters

Calculation of retinal roundness using a computer based unique technique

Poster Details

First Author: O.Inam TURKEY

Co Author(s): D. Somer   G. �zkan �ney   A. Soylu              

Abstract Details

Purpose:

The main aim is to quantify the retinal roundness (RR) at the 6500μm range of macula with a 100μm sampling rate at horizontal scans by using a novel computer-based application which we coded using Java interference.

Setting:

This study was approved by the Hacettepe University Faculty of Medicine Ethical Committee. The study was conducted at Hacettepe University Department of Biophysics as a Ph.D. Thesis Work.

Methods:

100 eyes of the 50 subjects were chosen for this retrospective study randomly. Horizontal images of SD-OCT (Heidelberg Engineering, Heidelberg, Germany) were exported as JPEG file and scale were set according to the original scale using Fiji and Java program. The program that we coded was initiated and 6500μm range centered to the fovea were chosen as for retina. Regions were divided into vertically 65 equal parts for 100μm and 3 horizontally equal parts (inner, middle, outer) furthermore and retinal roundness was calculated for each part.

Results:

There are 25 female and 25 male in the study. Mean age was 40.28±14.52 (16-62). Lowest horizontal RR was 0.09±0.00 (nasal-1700μm) and highest horizontal RR was 0.14±0.01 (temporal-200μm). Lowest inner, middle and outer RR was 0.27±0.02 (nasal-1500μm), 0.26±0.01 (nasal-1300μm), 0.26±0.01 (temporal-1000μm) and highest inner, middle and outer RR was 0.45±0.05 (nasal-200μm), 0.42±0.04 (central-0μm), 0.43±0.05 (central-0μm).

Conclusions:

Technology allows us to quantify the previously well-known situations. In this manner, Semi-automated computer-based programs can be very handy to find and define new indexes and parameters that can be used both diagnostic and evaluative purposes such in our case. RR can be a potential indicator for many ophthalmological and non-ophthalmological conditions but further investigations are needed.

Financial Disclosure:

None