Normative Percentiles of Retinal Nerve Fiber Layer Thickness and Glaucomatous Visual Field Loss.
Singh Rishabh, Rauscher Franziska G, Li Yangjiani, Eslami Mohammad, Kazeminasab Saber, Zebardast Nazlee, Wang Mengyu, Elze Tobias
AI Summary
This study found that OCT's normative percentiles of retinal nerve fiber layer thickness did not improve prediction of glaucomatous visual field loss compared to raw thickness measurements, challenging current clinical assumptions.
Abstract
Purpose
Circumpapillary retinal nerve fiber layer thickness (RNFLT) measurement aids in the clinical diagnosis of glaucoma. Spectral domain optical coherence tomography (SD-OCT) machines measure RNFLT and provide normative color-coded plots. In this retrospective study, we investigate whether normative percentiles of RNFLT (pRNFLT) from Spectralis SD-OCT improve prediction of glaucomatous visual field loss over raw RNFLT.
Methods
A longitudinal database containing OCT scans and visual fields from Massachusetts Eye & Ear glaucoma clinic patients was generated. Reliable OCT-visual field pairs were selected. Spectralis OCT normative distributions were extracted from machine printouts. Supervised machine learning models compared predictive performance between pRNFLT and raw RNFLT inputs. Regional structure-function associations were assessed with univariate regression to predict mean deviation (MD). Multivariable classification predicted MD, pattern standard deviation, MD change per year, and glaucoma hemifield test.
Results
There were 3016 OCT-visual field pairs that met the reliability criteria. Spectralis norms were found to be independent of age, sex, and ocular magnification. Regional analysis showed significant decrease in R2 from pRNFLT models compared to raw RNFLT models in inferotemporal sectors, across multiple regressors. In multivariable classification, there were no significant improvements in area under the curve of receiver operating characteristic curve (ROC-AUC) score with pRNFLT models compared to raw RNFLT models.
Conclusions
Our results challenge the assumption that normative percentiles from OCT machines improve prediction of glaucomatous visual field loss. Raw RNFLT alone shows strong prediction, with no models presenting improvement by the manufacturer norms. This may result from insufficient patient stratification in tested norms.
Translational relevance: Understanding correlation of normative databases to visual function may improve clinical interpretation of OCT data.
MeSH Terms
Shields Classification
Key Concepts4
Supervised machine learning models comparing predictive performance between normative percentiles of retinal nerve fiber layer thickness (pRNFLT) and raw RNFLT inputs showed no significant improvements in area under the curve of receiver operating characteristic curve (ROC-AUC) score with pRNFLT models compared to raw RNFLT models for predicting glaucomatous visual field loss.
Raw retinal nerve fiber layer thickness (RNFLT) alone shows strong prediction of glaucomatous visual field loss, with no models presenting improvement by the manufacturer norms (normative percentiles from OCT machines).
Regional analysis in a retrospective study showed a significant decrease in R2 from normative percentiles of retinal nerve fiber layer thickness (pRNFLT) models compared to raw RNFLT models in inferotemporal sectors, across multiple regressors, when assessing regional structure-function associations to predict mean deviation (MD).
Spectralis OCT normative distributions were found to be independent of age, sex, and ocular magnification in a retrospective study using a longitudinal database of OCT scans and visual fields from Massachusetts Eye & Ear glaucoma clinic patients.
Related Articles5
Utilization of Image-Based Deep Learning in Multimodal Glaucoma Detection Neural Network from a Primary Patient Cohort.
Observational StudyA Deep Learning Model for Glaucoma Detection Outperforms Conventional Metrics.
Observational StudyDiagnostic Performance of a Handheld Electroretinography Test for Glaucoma.
Cross-Sectional StudyDetecting Glaucoma from Fundus Photographs Using Deep Learning without Convolutions: Transformer for Improved Generalization.
Evaluation of a diagnostic technologyOCT Angiography Artifacts in Glaucoma.
Retrospective StudyIs this article assigned to the wrong chapter(s)? Let us know.