Functional analysis of glaucoma data

Area of Science:

• Ophthalmology
• Biostatistics
• Medical Data Analysis

Background:

• Glaucoma is an eye disorder characterized by elevated intraocular pressure (IOP), potentially damaging the optic nerve.
• IOP is a dynamic measure, varying over time, necessitating time-series analysis.
• Functional data analysis (FDA) offers a robust framework for analyzing time-dependent phenomena like IOP.

Purpose of the Study:

• To apply FDA to analyze IOP data from glaucoma patients.
• To identify key factors influencing IOP variability in glaucoma.
• To develop a predictive model for left eye IOP based on right eye IOP.

Main Methods:

• Utilized functional data analysis, specifically smoothed functional principal component analysis (SFPCA).
• Analyzed longitudinal IOP data from 35 right-eye glaucoma patients (2007-2011).
• Explored correlation patterns and developed a time-lagged regression model.

Main Results:

• Extracted main variation patterns and identified key factors contributing to glaucoma-related IOP changes.
• Revealed significant correlation patterns between right and left eye IOP.
• Established a model to predict left eye IOP using past right eye IOP measurements.

Conclusions:

• Functional data analysis effectively captures IOP dynamics in glaucoma.
• The developed model can aid in predicting contralateral eye IOP, potentially improving patient management.
• Understanding IOP variability is crucial for glaucoma diagnosis and treatment.