A Snapshot Adaptive Optics and Hyperspectral Autofluorescence Fundus Camera for Age-Related Macular Degeneration (AMD)

An imaging modality that allows for fast, simultaneous, noninvasive probing of both 3D cellular resolution retinal morphology by optical coherence tomography (OCT) and molecular-specific functions by autofluorescence (AF) could substantially improve both basic understanding and the early diagnosis of age-related macular degeneration (AMD), the leading cause of blindness in the developed world. The evaluation and management of AMD utilize several investigation modalities, but advancements in OCT technology have significantly contributed to better understanding of the disease, and have helped with monitoring progression and therapeutic efficacy. However, due to optical aberrations of the eye, the transverse resolution of conventional OCT is generally limited to 10-15 µm, restricting its use to visualize individual retinal cells in vivo. The integration of adaptive optics (AO) into OCT has demonstrated an immense success in mitigating these aberrations. Among various AO-OCT techniques, computation-based AO (CAO) becomes the spotlight of research because it shows unique advantages in data postprocessing flexibility and a reduced system cost. However, CAO is extremely sensitive to phase stability. The rapid motion of the eye can easily scramble the phase of reflected photons, restricting imaging to a single en-face layer.

To overcome this problem, the study team will integrate a snapshot hyperspectral imaging method, Image Mapping Spectrometry (IMS), with full-field spectral-domain OCT. The integrated system will first enable 3D CAO imaging of the retina because the single camera exposure (4 s),is too fast for eye movement to scramble phase between layers. Next, to improve resolution in 3D, the study team will adapt an established CAO algorithm to correct for wavefront aberrations. The resultant method, which the study team terms snapshot ultra-high-resolution OCT, will allow an acquisition of a quarter million A-scans simultaneously. Given a typical flash illumination duration (4us), the equivalent A-scan rate is 62.5 GHz, which is approximately three orders of magnitude faster than the state-of-the-art methods. Furthermore, to expand the system's functionality to molecular imaging, the study team will add a second IMS imaging channel for simultaneous hyperspectral imaging of retinal pigment epithelium (RPE) AF, enabling spectral biopsy of RPE and subRPE lesions such as drusen, the hallmark lesion of early AMD. The resultant dual-channel AO-OCT/AF system will be the first imaging modality that can provide both structural and molecular information about the retina in vivo and in 3D. The study team envisions such a system would shift the landscape of AMD evaluation and management. The insights so obtained will be of high value in clinical diagnosis and treatment. In addition, such a system will accelerate translational research with sensitive and early outcome testing of prospective therapeutic agents, saving sight and thereby providing enormous benefit to society.