Image Evaluation of Intalight Dream OCTA Scans With Optos FA & ICGA Images

1. Protocol Summary

The goal of this study is to evaluate the image quality and clinical utility of the OCTA images from the VG200D OCT device (investigational device) compared against the FA and ICGA images from the UWF Optos California (CA) device. Image assessment will be conducted by three expert image graders and involve the assessment of the images based on both image quality and clinical utility. Evidence for substantial equivalence will be obtained through establishing similar image evaluation results across the devices based on image quality and clinical utility.

Objectives and Endpoints

The primary objective is to evaluate image quality and clinical utility for visualization of the retinal vasculature from en-face OCTA scans between the VG200D and FA and ICGA images from the CA SLO device in eyes with retinal pathology.

The endpoint is the image grading results of the scans of the vascular structure of the retina in eyes with retinal associated pathology.

Methods

This is a prospective comparative study that will be conducted at one clinical site in the United States, in which subjects who sign an informed consent form and fulfil all inclusion and exclusion criteria will have images obtained using the investigational study device and the predicate device (VG200D and CA respectively).

This study will employ one VG200D device and one Optos CA device. At least one acceptable scan will be captured for each subject on each device. Scans will be assessed at the time of capture by the operator for acceptable quality, poor quality scans will be retaken. All scans will be saved and both acceptable and unacceptable scans will be recorded on the CRF. For FA and ICGA scans, the CA device captures multiple images over the time-course of the dye filling the arteries and veins, however the operator and/or investigator will review and choose the most clinically appropriate image to use for evaluation.

Subjects enrolled in this study will be confirmed to have retina associated pathology at the time of the study visit and will be in need of an FA and/or ICGA scan. Ocular status will be confirmed by a clinical examination during the study visit. Subjects enrolled may have more than one ocular pathology. Acceptable retina associated pathologies may include but are not limited to: Age-related macular degeneration, Diabetic Retinopathy with or without DME, and/or Vein or artery occlusions, that require FA or ICGA imaging.

If both eyes qualify, the study eye will be selected by the Investigator based on retinal pathology, where the study eye will have more significant pathology. If only one eye qualifies, that eye will be the study eye. Device imaging order will be randomized.

All scans are reviewed for acceptable image quality by the operator at the time of image acquisition based on user manual described criteria and as described in section 7.1.1 below. Operators will use the pre-defined, objective criteria described in section 7.1.1 below for how images are included into or excluded from the final data set and will be applied by the device operators using pre-defined image quality parameters, such as the signal strength (for OCTA images), poor illumination or contrast (for SLO images), presence of blinks, eye-movements, clipping, etc. These criteria will be applied to all scans at the time of capture by the operator and/or investigator. Acceptable scans will solely be determined by the operator and investigator. The number of scans that are rejected for all devices will be recorded in the CRF and the reasons for rejection will be provided. Image accountability of all scans will be reported in the final clinical study report.

The VG200D has more than one OCTA scan pattern available and all OCTA scan patterns for the VG200D will be captured for image evaluation. The first acceptable OCTA scan from each scan pattern captured will be used for the en-face OCTA visualization assessment.

The en-face OCTA images will be assessed and graded by three experienced graders and scored for image quality and clinical utility as compared to a similar grading of the Optos CA FA and ICGA images.

Image Grading

OCTA en-face images and FA and ICGA images will be exported and evaluated by three masked graders in a masked and randomized fashion for image quality and clinical utility. Image graders will have experience in both OCTA images and FA and ICGA images and will be independent from the clinical research site operator and investigator. Grading will be masked and randomized. Graders will work independently evaluating a single image at a time for the FA and/or ICGA images and will review all OCTA depth slabs for the OCTA scans. For OCTA scans, where multiple images at various depths are available, all depth scans (slabs) will be reviewed together for that patient and an overall image assessment will be provided (i.e., a Gestalt assessment will be made based on all images).

All graders will score each image for general image quality on a five-point scale from 0-5 where scores of 0 indicate image failure, 1 indicates very poor image quality, 2 indicates poor image quality, 3 indicates average image quality, 4 indicates good image quality, and 5 indicates excellent image quality. Scores of 0, 1 and 2 indicate the images have unacceptable image quality, and scores of 3, 4 and 5 indicate the images have acceptable image quality. In addition, images will be graded for clinical utility in helping to aid clinical decisions, such as viewing normal structures (confirming absence of pathology) or detecting pathological changes in the images (confirming pathology). The graders will evaluate clinical utility based on image content (e.g., identifying key vascular features such as the foveal avascular zone border, large-, medium-, and small-sized blood vessels, and blind-end capillaries; key pathologic vascular features such as microaneurysms, neovascular networks, telangiectasias, capillary dropout or nonperfusion, etc. En face depth sections (thickness slabs) from all available depths will be assessed together. The clinical utility or usefulness of OCTA images will also be graded on a five-point scale from 0-5 where scores of 0 indicate no clinical information is present, 1 indicates clinical information is very poor and not helpful, 2 indicates clinical information is poor and not helpful, 3 indicates clinical information is average but helpful, 4 indicates clinical information is good and helpful, and 5 indicates the clinical information present is excellent and very helpful. Scores of 0, 1 and 2 indicate the images are not clinically helpful and are considered unacceptable for providing useful clinical information. Scores of 3, 4 and 5 indicate the images are adding useful clinical information for aiding clinical decisions and are acceptable.

Planned Analyses

Results will be analyzed for differences using the non-parametric Wilcoxon signed rank test. In addition, the percentage agreement between each pair of readers will be determined by using 2x2 tables for the binary outcomes of acceptable vs. not acceptable.

Analysis of Safety Any adverse event (AE) associated with the VG200D or the predicate devices will be used to evaluate safety. Adverse event definitions and reporting requirements are provided in the study Protocol. All adverse events will be reported in the clinical study report.