Purchase this article with an account.
Mariel S. Lavieri, Sam Devaprasad, Manqi Li, Greggory Schell, Pamela Martinez Villarreal, Jonathan E Helm, Mark Van Oyen, David C Musch, Joshua D Stein; User Friendly Tool Using Kalman Filter Algorithms to Display Glaucoma Progression Indicators and Personalized Time to Next Test. Invest. Ophthalmol. Vis. Sci. 2014;55(13):5611. doi: https://doi.org/.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To develop a user friendly interface that can visualize measurements obtained from tonometric and perimetric testing of patients with open angle glaucoma (including intraocular pressure (IOP), mean deviation (MD), and pattern standard deviation (PSD)), and to display personalized, dynamically-adjusted forecasts of these indicators and suggestions for the timing of future testing.
Kalman filtering algorithms are used to model the IOP, MD and PSD dynamics of patients with Open Angle Glaucoma (OAG) and to update the knowledge about those dynamics as additional readings are obtained. We developed a tool in Excel using Visual Basic for Applications (VBA). It runs MATLAB code that generates forecasted values for IOP, MD, and PSD for each individual patient as well as suggested testing frequencies. The tool also displays data from past visits and changes in visual field readings to enhance clinical decision making. This tool has been calibrated and validated using longitudinal data from patients who were enrolled in Collaborative Initial Glaucoma Treatment Study (CIGTS) and the Advanced Glaucoma Intervention Study (AGIS).
A sample screenshot from the tool is presented in Figure 1. This tool can calculate adjusted predictions and personalized testing frequencies in seconds, providing clinicians with valuable insight to enable personalized glaucoma management.
It is possible to integrate a Kalman Filter approach into a user-friendly tool that improves visualization of tonometric and perimetric measurements and how they change over time. This tool generates real-time personalized predictions of the status of each patient’s OAG and when future testing should be performed so as to not miss progression.
This PDF is available to Subscribers Only