Circular functional analysis of OCT data for precise identification of structural phenotypes in the eye

Abstract

Progressive optic neuropathies such as glaucoma are major causes of blindness globally. Multiple sources of subjectivity and analytical challenges are often encountered by clinicians in the process of early diagnosis and clinical management of these diseases. In glaucoma, the structural damage is often characterized by neuroretinal rim (NRR) thinning of the optic nerve head, and other clinical parameters. Baseline structural heterogeneity in the eyes can play a key role in the progression of optic neuropathies, and present challenges to clinical decision-making. We generated a dataset of Optical Coherence Tomography (OCT) based high-resolution circular measurements on NRR phenotypes, along with other clinical covariates, of 3973 healthy eyes as part of an established clinical cohort of Asian Indian participants. We introduced CIFU, a new computational pipeline for CIrcular FUnctional data modeling and analysis. We demonstrated CIFU by unsupervised circular functional clustering of the OCT NRR data, followed by meta-clustering to characterize the clusters using clinical covariates, and presented a circular visualization of the results. Upon stratification by age, we identified a healthy NRR phenotype cluster in the age group 40–49 years with predictive potential for glaucoma. Our dataset also addresses the disparity of representation of this particular population in normative OCT databases.

Focused analysis of RNFL decay in glaucomatous eyes using circular statistics on high-resolution OCT data

“…While past studies have used spatial, trigonometric and Fourier analysis on OCT data, e.g., [25][26][27][28], our framework, involving directional or circular data, is methodologically quite different from those approaches. While our earlier platform CIFU (14) focuses on the aspects of shapes and curves in OCT NRR data, as in functional data analysis (FDA), it does not explicitly address the directional characteristics of OCT data or angular measurements thereof. We have demonstrated the utility of both the granularity and the directionality of OCT data and combined these with an angle-based description of patterns that could be studied with innovative downstream analyses.…”

“…In addition to age, on the structural side, biological heterogeneity of ONH phenotypes, with or without any neuropathy, can also present challenges to clinical decision-making. Unsupervised learning of the heterogeneity of normative ONH phenotypes in a given population can provide key insights into the diversity of baselines that may exist for degenerative neuropathies, and thus, reduce subjectivity in diagnosis [ 14 ]. Such knowledge is particularly useful in glaucoma for which different ONH parameters play a combined role in early detection.…”