Quantifying Discordance Between Structure and Function Measurements in the Clinical Assessment of Glaucoma

Abstract

To evaluate a new method of quantifying and visualizing discordance between structural and functional measurements in glaucomatous eyes by predicting the visual field (VF) from retinal nerve fiber layer thickness (RNFLT) using a bayesian radial basis function. Methods: Five GDx VCC RNFLT scans and 5 Humphrey 24-2 Swedish Interactive Thresholding Algorithm VF tests were performed for 50 glaucomatous eyes from 50 patients. A best-available estimate (BAE) of the true VF was calculated as the pointwise median of these 5 replications. This BAE VF was compared with every RNFLT-predicted VF from the bayesian radial basis function and every measured VF. Predictability of VFs from RNFLT was established from previous data. A structure-function pattern discordance map and a structure-function discordance index (scores of 0-1) were established from the predictability limits for each structurefunction measurement pair to quantify and visualize the discordance between the structure-predicted and measured VFs. Results: The mean absolute difference between the structure-predicted and BAE VFs was 3.9 dB. The mean absolute difference between measured and BAE VFs was 2.6 dB. The mean (SD) structure-function discordance index score was 0.34 (0.11). Ninety-seven (39%) of the structure-predicted VFs showed significant discordance (structure-function discordance index score Ͼ0.3) from measured VFs. Conclusions: On average, the bayesian radial basis function predicts the BAE VF from RNFLT slightly less well than a measured VF from the 5 VFs composing the BAE VF. The pattern discordance map highlights locations with structurefunction discordance, with the structure-function discordance index providing a summary index. These tools may help clinicians trust the mutually confirmatory structurefunction measurements with good concordance or identify unreliable ones with poor concordance.

Does the Swedish Interactive Threshold Algorithm (SITA) accurately map visual field loss attributed to vigabatrin?

“…The grey scale plots from all 3 testing strategies are presented for every patient with confirmed VGB attributed field loss (Figure 3 ) and also those without visual field loss (Additional file 2 : Figure S1). Visual field plots were generated from the Humphrey sensitivity values (dB) using software which is freely available from one of the authors (HZ) [ 23 , 24 ]. Visual inspection of the plots shows they were approximately equivalent, for all 3 algorithms, for every patient with and without a confirmed VGB attributed visual field defect.…”

Does the Swedish Interactive Threshold Algorithm (SITA) accurately map visual field loss attributed to vigabatrin?

“…The grey scale plots from all 3 testing strategies are presented for every patient with confirmed VGB attributed field loss (Figure 3 ) and also those without visual field loss (Additional file 2 : Figure S1). Visual field plots were generated from the Humphrey sensitivity values (dB) using software which is freely available from one of the authors (HZ) [ 23 , 24 ]. Visual inspection of the plots shows they were approximately equivalent, for all 3 algorithms, for every patient with and without a confirmed VGB attributed visual field defect.…”

Estimating the Rate of Retinal Ganglion Cell Loss in Glaucoma

“…This highlights the need for a combined approach using structure and function to detect disease progression. 24–26 The ability to express results of functional and structural tests in the same domain opens the possibility of combining the information from the two tests to increase the precision of RGC estimates, as performed in our study. By combining the estimates, one increases the precision of the final estimate of neuronal losses to better detect change over time.…”

Predicting Visual Fields From Optical Coherence Tomography via an Ensemble of Deep Representation Learners

“…One goal is for our model to provide a relevant clinical tool that indicates concordance between the VF and the chosen surrogate measure for structural loss. For instance, when a VF and a structural measure are available, a chart mapped in VF space could be provided indicating areas where the measurements are in concordance (within a certain range of accuracy and precision) and where they are not [52]. This chart could provide clinically useful information about the diagnosis or the reliability of the individual measurements.…”

Does the Swedish Interactive Threshold Algorithm (SITA) accurately map visual field loss attributed to vigabatrin?

“…The grey scale plots from all 3 testing strategies are presented for every patient with confirmed VGB attributed field loss (Figure 3 ) and also those without visual field loss (Additional file 2 : Figure S1). Visual field plots were generated from the Humphrey sensitivity values (dB) using software which is freely available from one of the authors (HZ) [ 23 , 24 ]. Visual inspection of the plots shows they were approximately equivalent, for all 3 algorithms, for every patient with and without a confirmed VGB attributed visual field defect.…”

Estimating the Rate of Retinal Ganglion Cell Loss in Glaucoma

“…This highlights the need for a combined approach using structure and function to detect disease progression. 24–26 The ability to express results of functional and structural tests in the same domain opens the possibility of combining the information from the two tests to increase the precision of RGC estimates, as performed in our study. By combining the estimates, one increases the precision of the final estimate of neuronal losses to better detect change over time.…”

Predicting Visual Fields From Optical Coherence Tomography via an Ensemble of Deep Representation Learners

“…One goal is for our model to provide a relevant clinical tool that indicates concordance between the VF and the chosen surrogate measure for structural loss. For instance, when a VF and a structural measure are available, a chart mapped in VF space could be provided indicating areas where the measurements are in concordance (within a certain range of accuracy and precision) and where they are not [52]. This chart could provide clinically useful information about the diagnosis or the reliability of the individual measurements.…”

Does the Swedish Interactive Threshold Algorithm (SITA) accurately map visual field loss attributed to vigabatrin?

“…The grey scale plots from all 3 testing strategies are presented for every patient with confirmed VGB attributed field loss (Figure 3 ) and also those without visual field loss (Additional file 2 : Figure S1). Visual field plots were generated from the Humphrey sensitivity values (dB) using software which is freely available from one of the authors (HZ) [ 23 , 24 ]. Visual inspection of the plots shows they were approximately equivalent, for all 3 algorithms, for every patient with and without a confirmed VGB attributed visual field defect.…”

Estimating the Rate of Retinal Ganglion Cell Loss in Glaucoma

“…This highlights the need for a combined approach using structure and function to detect disease progression. 24–26 The ability to express results of functional and structural tests in the same domain opens the possibility of combining the information from the two tests to increase the precision of RGC estimates, as performed in our study. By combining the estimates, one increases the precision of the final estimate of neuronal losses to better detect change over time.…”

Predicting Visual Fields From Optical Coherence Tomography via an Ensemble of Deep Representation Learners

“…One goal is for our model to provide a relevant clinical tool that indicates concordance between the VF and the chosen surrogate measure for structural loss. For instance, when a VF and a structural measure are available, a chart mapped in VF space could be provided indicating areas where the measurements are in concordance (within a certain range of accuracy and precision) and where they are not [52]. This chart could provide clinically useful information about the diagnosis or the reliability of the individual measurements.…”