Abstract
Purpose :
Incomplete blinking and reduced maximum blink interval (MBI) are characteristics of dry eye disease (DED). Blinking also impacts the rheology and retention of artificial tear formulations (ATFs) on the ocular surface. In this study, we have tested ADS1299, a high-precision 24-bit A/D for EEG, to capture blink profiles as vertical EOGs with precision and speed.
Methods :
Active electrodes, placed above and below the eyes, were connected to inputs of an embedded system for ADS1299 with in-built amplifiers and filters (ADS1299EEGFE-PDK; Texas Instruments). The reference electrode, positioned below the L/R ear, was connected to the ground. EOGs were recorded at 250 Hz for 1-5 min, uploaded to a PC via USB, and filtered digitally to remove slow drifts and high-frequency noise. Subsequent thresholding enabled the identification of blink boundaries. The segmented blinks were analyzed to determine blink amplitude, velocities, and interblink interval. The MBI was assessed over 2 minutes, during which subjects were instructed to keep their eyes open as long as possible. This extended MBI protocol enables repeat measurements of MBI.
Results :
EOG was noise-free but showed slow drifts and was affected by non-blink-associated eye movements. The non-blink effects on the recordings could be removed during post-processing. The segmented blinks could be validated by video recording using a USB camera (60 fps). In > 50 healthy eyes, the interblink interval, max opening/closing velocities of the eyelid, and amplitude of the blinks were comparable to those in previous reports. EOG gives an accurate estimate of MBI, which is prolonged in healthy subjects but shorter and variable in DED.
Conclusions :
ADS1299 is suitable for high-precision EOG, but ADS1299EEGFE-PDK could not be used to synchronize the recordings with data acquisition from other instruments (e.g., ocular fluorometer) or to obtain long-term recordings. Nonetheless, our studies show that fast-sampling EOG can effectively assess MBI.
This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.