Abstract
Currently, there are about 40 to 60 million Americans suffering from Dry Eye Syndrome (DES); this serious public health problem will worsen with the explosive aging population created by baby boomers, where DES has a high incidence. However, the therapeutics for DES are elusive because our understanding of DES is so elementary, especially the correlation between symptoms and the diagnosis. Unfortunately, a quantitative diagnosis, which is the prerequisite to advance the management of DES, is yet to be realized. We are seeking the next breakthrough in DES management by providing a quantitative diagnosis, with the combination of optical coherence tomography (OCT) imaging and statistical decision theory. The statistical decision theory is formulated for the specific task of the tear film imaging, thus it will allow the assessment and optimization of the OCT system for such an application. Furthermore, the combination of the statistical decision theory and OCT will show its advantage by taking into account the system noise, compared to the direct measurements from the imaging itself. We present the mathematical models of the OCT system and the decision-making mechanisms. We first apply the receiver operating curve (ROC) analysis for a detection task and explore the limit of the OCT imaging system. Then we implement a maximum-likelihood estimator for the quantification of the tear film thickness; the estimator can estimate thinner thickness than the axial resolution of the OCT system, which is beyond the limitation of direct measurements from OCT images.
Meeting abstract presented at OSA Fall Vision 2012