March 9, 2018 Conference


Emerging Novel Automatic Image Analysis to Allow Early Detection of Systemic Diseases based upon OCT and OCTA Imaging

Dr. Shlomit Schaal (Presenter)

Abstract Title:

Emerging Novel Automatic Image Analysis to Allow Early Detection of Systemic Diseases based upon OCT and OCTA Imaging

Author Name:

Shlomit Schaal


Department of Ophthalmology and Visual Sciences

University of Massachusetts Medical School. Worcester, MA


To develop and employ a novel mathematical software algorithm that enables the automatic detection of early microvasculature changes in the retina and to correlate these automatically detected changes with the presence of systemic diseases.

This is an ongoing a prospective, observational study that includes more than 500 patients with systemic diseases (sleep apnea, diabetes, hypertension, and pre-eclampsia) and age matched controls. Optical coherence tomography angiography (OCTA) was performed using Cirrus HD-OCT 5000 Angioplex (Carl Zeiss Meditech, CA. USA). Patients underwent 3x3mm and 6x6mm macular scans that were captured at ~840nm wavelength and 68,000 A-scans/second and the split-spectrum amplitude-decorrelation angiography algorithm was utilized. An automatic mathematical analysis software was developed to automatically detect retinal microvasculature on OCT and on OCTA images, including superficial and deep retinal cuts. The developed software consisted of three main stages: firstly reduce the noise and improve the contrast by using the GGMRF model, secondly retinal segmentation was performed by integrating current and prior intensity models, and a higher-order spatial MGFR model. Finally, total retinal microvasculature analysis was performed by applying connectivity analysis to present more accurate results.

The automatic OCT and OCTA analysis software detected early development of micro-vascular changes that were not apparent by other investigation modalities as well as areas of capillary loss. Alterations in vascular structure were noted as well as increased vessel and capillary tortuosity. Enlargement of the foveal avascular zone (FAZ) appeared to be one of the earliest changes in diabetic patients. The automatic software was able to accurately detect areas of non-perfusion and decreased vessel density. Automatic detection of early microvasculature pathology carries the promise of early detection of systemic diseases to allow prevention of progression towards more advanced retinopathy.