Use of a Retinal Phantom to Facilitate Adaptive Optics Scanning Light Ophthalmoscopy Imaging in a Multicenter Clinical Trial.
Researchers
Joseph Kreis, Jessica Wong, Yushan Yang, Palash Bharadwaj, Robert F Cooper, Mina Gaffney, Nivedhitha Govindasamy, Amy L Green, Brian P Higgins, Yu You Jiang, Angelos Kalitzeos, Michel Michaelides, Jessica I W Morgan, Jason Porter, Jeremy D Rogers, Ian Rosenthal, Ramkumar Sabesan, Kaitlyn A Sapoznik, Alexander W Schill, Ryan D Sochol, Yiyi Wang, Raymond L Warner, Emma Warr, Benjamin J Wendel, Niamh Wynne, Peiluo Xu, Sergey Tarima, Daniel X Hammer, Anant Agrawal, Jacque L Duncan, Joseph Carroll
Abstract
To describe the use of a model eye-mounted retinal phantom to support adaptive optics scanning light ophthalmoscope (AOSLO) imaging in a multicenter clinical trial (NCT05537220). Six three-dimensional-printed retinal phantoms were mounted in model eyes. These phantoms contain equally spaced arrays of reflective "cone-like" structures mimicking foveal cone photoreceptor spacing. Intrasession repeatability of phantom imaging was assessed for all six phantoms on two custom AOSLOs. One phantom was also imaged on two additional AOSLOs to assess interdevice reproducibility. Lastly, each of the six phantoms was imaged longitudinally at their respective trial sites. Intrasession repeatability of phantom-derived horizontal "cone" spacing measures was high for two different AOSLOs. However, intersystem differences were found in the average horizontal "cone" spacing of five of eight patches assessed. Across four unique AOSLOs, intersystem differences were found in the average horizontal "cone" spacing of seven of eight patches assessed. A linear regression analysis of longitudinal phantom imaging with a single AOSLO revealed few significant changes in horizontal "cone" spacing over the course of about 2 years. Similar results were observed for vertical "cone" spacing measures across the three experiments. Longitudinal spacing measurements were generally stable in our study. However, small but significant differences in spacing were observed between systems. The approach described here may facilitate combining cone metrics extracted from images acquired using different AOSLOs in multicenter studies. Cross-system validation strategies can support the use of AOSLO imaging in multicenter clinical trials.Source: PubMed (PMID: 42635483)View Original on PubMed