Patient fixation analysis in a combined OCT and fundus imaging system.

Abstract

Purpose : The quality of fixation achieved can greatly improve the OCT image acquisition by reducing eye motion. The patient fixation analysis serves as a measure of feedback during and after acquisition. This assessment can then be used during or after the scan to remind the patient to fixate better. In this work we demonstrate that meticulous patient fixation analysis can improve the acquisition workflow.

Methods : The motion of the eye is determined by calculating movement of the retina in the infrared-reflectance (IR) fundus image that appears in a sequence. IR images are acquired by using an IR fundus imager which is a separate sub-system from the OCT. This motion can either be horizontal (x) or vertical (y) direction along with some minor rotation. IR tracking prototype software of the CLARUS^TM 500 (Zeiss, Dublin, CA) computes the tracking parameters of x and y eye motion relative to a fixation position. We analyzed the overall eye motion during an OCT image acquisiton using these tracking parameters. Each eye was scanned using 3 different motion levels: good fixation, systematic eye movement, and random eye movement. The mean and standard deviation of the eye motion were reported for each motion level and each scan.

Results : Around 500 CLARUS IR images per scan (768x624 pixels over 11.52x9.36 mm² at the frame rate of 50Hz) of 15 eyes were captured with induced eye motion. Figure 1 shows the x and y eye motion relative to the inital fixation for three subjects with different induced eye motion. Figure 2 shows the statistics of eye motion for each scan of the 15 subjects. The mean value indicates the overall fixation offset from the initial fixation position. The standard deviation indicates a measure of overall eye motion within an acquisition. Scans containing systematic or random eye movement show significantly greater mean and standard deviation compared to good fixation, which is used as an indicator for poor fixation.

Conclusions : We track the retina and perform fixation analysis to highlight its use as feedback for the operator or the patient by providing informative messages during or after acquisiton for reduced motion in OCT images, which is important for subsequent data processing.

This is a 2021 ARVO Annual Meeting abstract.

 

Figure 1: x and y translations for systematic, random movement and good fixation for 3 different acquisitons have been overlayed on the same plot.

 

Figure 2: The statistics of eye motion for each scan.