Advanced Imaging for Glaucoma Study

Glaucoma is a leading cause of blindness in the US. Traditional methods of glaucoma diagnosis and monitoring lack good sensitivity and specificity. Delays in detecting glaucoma progression can lead to inadequate treatment and irreversible visual loss. Our goal is to improve glaucoma diagnosis by utilizing new imaging modalities that can reveal changes in the retinal layers affected by glaucoma and the associated reduction in retinal blood flow. Glaucoma selectively damages the retinal nerve fibers, which originate from cell bodies in ganglion cell layer (GCL) and travel to the optic nerve via the nerve fiber layer (NFL). We hypothesize that subtle damages in these structures can be detected earlier by optical coherence tomography (OCT) and other advanced imaging modalities than with current standard methods. OCT is based on infrared light reflectometry. It provides micrometer-scale cross-sectional images of retinal structures, which are not possible with other non-invasive techniques. More than 7,000 OCT systems are already being used for the diagnosis of glaucoma and retinal diseases. Phase I of the Advanced Imaging for Glaucoma (AIG) study demonstrated that peripapillary NFL thickness measured with the standard timedomain (TD) OCT technology has higher glaucoma diagnostic accuracy than other quantitative diagnostic technologies such as scanning laser polarimetry (SLP) and scanning laser tomography (SLT). We also demonstrated that more advanced diagnostic software and faster Fourier-domain (FD) OCT systems can achieve even better diagnostic accuracy and reproducibility. In the proposed Phase II of the AIG study, we will continue the most promising aspects of the research to further improve both technology and clinical practice. The AIG Partnership investigators at the Oregon Health \& Science University (OHSU), Massachusetts Institute of Technology (MIT), and University of Pittsburgh (UP) include those who invented OCT and pioneered its applications to glaucoma. OHSU, University of Southern California (USC), UP and University of Miami (UM) also have major glaucoma referral centers. The Partnership combines engineers and clinicians who have the track record and synergy to develop novel technologies, evaluate them in a rigorous clinical study, and transfer the knowledge to industry and medicine. The Specific Aims of this competing renewal proposal are: 1. Develop image processing and diagnostic analysis for 3-dimensional OCT data. The AIG study is currently using 26 kHz (axial scan repetition rate) FD-OCT technology that is capable of scanning the macula and the optic nerve head in a fraction of a second. We have completed computer algorithms for mapping and analysis of the macular ganglion cell complex (mGCC) and the peripapillary NFL, which lead to significant improvement in diagnostic accuracy. We propose to continue the work on disc cupping analysis, NFL reflectivity analysis, and expert system combination of multiple anatomic parameters to further improve diagnostic accuracy. Algorithms to detect progression of glaucoma over time are also planned. 2. Develop ultrafast OCT systems for imaging of the macula and optic nerve head. Although current FD-OCT technology at 26 kHz represents a tremendous advance over standard 400 Hz TD-OCT (Zeiss Stratus), it still takes \~4 seconds for a full 3-dimensional (3D) raster scan of the macula. Our goal is to reduce this time to 0.1-0.2 second so 3D scans will be minimally affected by eye movement. This requires an ultrafast speed of 500-1000 kHz. We plan to adapt the Fourier-domain modelocked-laser (FDML) swept-source OCT, which has already been demonstrated at 249 kHz at MIT. We will further improve its speed to 500 kHz. The short integration time and phase stability of FDML-OCT is ideal for Doppler perfusion measurement (see next aim). For an even faster speed, parallel line-scan FD-OCT at 1 MHz will be developed. Line-scan OCT is not suitable for Doppler flow measurement due to the relatively long integration time, but is more efficient for ultrafast anatomic imaging. It will allow full 8x8 mm macular 3D imaging in 0.2 second. We will also continue to develop polarization-sensitive (PS) OCT for NFL birefringence measurement, which will also be greatly enhanced by higher speed and greater averaging to suppress noise. 3. Develop Doppler OCT to measure retinal perfusion. One of the significant achievements of the AIG project is the demonstration of a reproducible method of measuring total retinal blood flow using Doppler FD-OCT. Reduced flow was found in glaucomatous eyes, opening an important new approach to measure the severity of glaucoma and assess the risk for further progression. An automated algorithm will be developed to improve the robustness of Doppler flow measurement. We will also investigate Doppler OCT with the ultrafast FDML-OCT system. 4. Evaluate OCT technologies in a longitudinal clinical study. An extension of the ongoing clinical study is proposed. Participants (1000 planned with 700+ already enrolled) in normal, glaucoma suspect, and glaucoma groups will be followed. OCT and other imaging technologies will be compared for diagnostic accuracy, detection of early progression, and prediction of future visual field loss. The impact of intraocular pressure on retinal blood flow and how flow affects the risk of glaucoma will also be studied. Quantitative imaging technologies such as OCT have improved glaucoma management by reducing reliance on insensitive tests such as perimetry and subjective disc grading. The AIG Partnership comprises engineers and clinicians who co-invented OCT. We propose to further improve its performance with higher speed, more sophisticated software, and novel functional measurements. The eventual goal is to save vision by basing glaucoma treatment decisions on speedy and reliable imaging tests.