Purchase this article with an account.
Taylor Shagam, Simon Antonio Bello, Sophie Kubach, Luis De Sisternes, Thomas Callan, Noli Graves, Roger A. Goldberg; Effects of OCTA scanning speed on image quality and clinical workflow. Invest. Ophthalmol. Vis. Sci. 2020;61(9):PB0093.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
In optical coherence tomography (OCT), it is well known that there is a trade-off between speed and scan quality. However, for OCT Angiography (OCTA), the trade-off has not yet been fully characterized, since the impact of speed on tracking and angiographic signal may affect overall quality. Eye care practitioners require good quality scans as well as a reasonable workflow. We conducted a study using a dual speed system to characterize the image quality and speed of acquisition as a function of A-scan speed.
A modified, dual speed, PLEX® Elite 9000 OCTA system (ZEISS, Dublin, CA) was used to image eyes with retinal pathology. Volunteers were divided into two groups and imaged using the following protocol:3x3 mm (300x300), 6x6 mm (500x500), 9x9 mm (500x500), 12x12 mm (500x500), and 15x9 mm (500x834).Group 1 was imaged with an OCT light source with an A-scan rate of 100 kHz, while group 2 acquisitions were made with an A-scan rate of 200 kHz. In all cases eye motion tracking was active, which re-scans any B-scan that presents motion >30 µm. Acquisition times were calculated by monitoring the scanning galvo electrical signals. A subject matter expert graded the superficial, deep and outer retina-choriocapillaris (ORCC) enface retinal slabs of each of these scans using a 1-5 scale. Acquisition times and image quality were then compared between the two groups. Statistical significance was established using a two-sample t-test with an α=0.05.
A total of 25 eyes were imaged on each group. On average, we observed a 30-50% improvement in acquisition time, depending on the scan pattern. Larger scan patterns, which historically showed longer acquisition times, present the biggest improvement. 96% of all data acquired at 200kHz was captured in <25 seconds, while only 64% of data acquired at 100 kHz met this criterion. Similarly, images acquired at 200 kHz showed improved imaged quality, with grades ranging from 3 to 4.5 out of 5. For all patterns, grades were significantly higher when acquired at 200 kHz versus 100 kHz.
Our study shows that despite reduced light exposure times, increasing the scanning speed from 100 to 200 kHz yields higher quality OCTA en face images. Similarly, the increase in scanning speed significantly improves clinical workflows by reducing the imaging time, possibly allowing clinical personnel to image a higher number of patients.
This is a 2020 Imaging in the Eye Conference abstract.
This PDF is available to Subscribers Only