Eye Disorders and Treatment

Biography

Biography: Lyne Racette

Abstract

While the presence and rate of glaucoma progression influences treatment decisions, the methods currently available to detect and monitor progression are imprecise and do not allow clinicians to make accurate assessments of the status of their patients. We have developed a dynamic structure-function (DSF) model to detect glaucoma progression, with the long-term goal of improving the detection and monitoring of glaucoma progression using structural and functional data jointly. This DSF model has two descriptors: Centroids and velocity vectors. The centroids represent the current stage of the disease, while the velocity vectors represent the trend of change of the centroids over time. The velocity vectors are in 2-dimensional space and show the direction (vector orientation) and rate (vector length) of change. We have also developed an individualized inferential component for our DSF model that is based on permutation analysis and that allows the model to tease out true progression from variability for each eye. Longitudinal data from patients with ocular hypertension or primary open-angle glaucoma enrolled in the diagnostic innovations in glaucoma study or in the African descent and glaucoma evaluation study were used to assess the prediction accuracy of the model. We used mean sensitivity as a functional measure and rim area as a structural measure. In this paper, we present the details of our DSF model, compare its prediction accuracy to that of ordinary least square linear regression, and assess its agreement with other measures of glaucoma progression.