Global Search

Search articles, concepts, and chapters

Am J OphthalmolMarch 20220 citations

The three-dimensional structural configuration of the central retinal vessel trunk and branches as a glaucoma biomarker.

Panda Satish K, Cheong Haris, Tun Tin A, Chuangsuwanich Thanadet, Kadziauskiene Aiste, Senthil Vijayalakshmi, Krishnadas Ramaswami, Buist Martin L, Perera Shamira, Cheng Ching-Yu


AI Summary

This study found that the 3D configuration of central retinal vessels, analyzed by AI, is a superior glaucoma diagnostic marker compared to standard RNFL thickness, potentially improving early detection.

Abstract

Purpose

To assess whether the 3-dimensional (3D) structural configuration of the central retinal vessel trunk and its branches (CRVT&B) could be used as a diagnostic marker for glaucoma.

Design

Retrospective, deep-learning approach diagnosis study.

Methods

We trained a deep learning network to automatically segment the CRVT&B from the B-scans of the optical coherence tomography (OCT) volume of the optic nerve head. Subsequently, 2 different approaches were used for glaucoma diagnosis using the structural configuration of the CRVT&B as extracted from the OCT volumes. In the first approach, we aimed to provide a diagnosis using only 3D convolutional neural networks and the 3D structure of the CRVT&B. For the second approach, we projected the 3D structure of the CRVT&B orthographically onto sagittal, frontal, and transverse planes to obtain 3 two-dimensional (2D) images, and then a 2D convolutional neural network was used for diagnosis. The segmentation accuracy was evaluated using the Dice coefficient, whereas the diagnostic accuracy was assessed using the area under the receiver operating characteristic curves (AUCs). The diagnostic performance of the CRVT&B was also compared with that of retinal nerve fiber layer (RNFL) thickness (calculated in the same cohorts).

Results

Our segmentation network was able to efficiently segment retinal blood vessels from OCT scans. On a test set, we achieved a Dice coefficient of 0.81 ± 0.07. The 3D and 2D diagnostic networks were able to differentiate glaucoma from nonglaucoma subjects with accuracies of 82.7% and 83.3%, respectively. The corresponding AUCs for the CRVT&B were 0.89 and 0.90, higher than those obtained with RNFL thickness alone (AUCs ranging from 0.74 to 0.80).

Conclusions

Our work demonstrated that the diagnostic power of the CRVT&B is superior to that of a gold-standard glaucoma parameter, that is, RNFL thickness. Our work also suggested that the major retinal blood vessels form a "skeleton"-the configuration of which may be representative of major optic nerve head structural changes as typically observed with the development and progression of glaucoma.


MeSH Terms

BiomarkersGlaucomaHumansIntraocular PressureROC CurveRetinal VesselsRetrospective StudiesTomography, Optical Coherence

Key Concepts5

The 3D diagnostic network using the 3D structure of the central retinal vessel trunk and its branches (CRVT&B) differentiated glaucoma from non-glaucoma subjects with an accuracy of 82.7% and an AUC of 0.89.

DiagnosisCohortRetrospective, deep-learning approach diagnosis studyn=Not specifiedCh5Ch11

The 2D diagnostic network, using orthographically projected 3D structures of the central retinal vessel trunk and its branches (CRVT&B) onto sagittal, frontal, and transverse planes, differentiated glaucoma from non-glaucoma subjects with an accuracy of 83.3% and an AUC of 0.90.

DiagnosisCohortRetrospective, deep-learning approach diagnosis studyn=Not specifiedCh5Ch11

The diagnostic performance of the central retinal vessel trunk and its branches (CRVT&B) using both 3D (AUC 0.89) and 2D (AUC 0.90) approaches was superior to that of retinal nerve fiber layer (RNFL) thickness alone (AUCs ranging from 0.74 to 0.80) for glaucoma diagnosis.

Comparative EffectivenessCohortRetrospective, deep-learning approach diagnosis studyn=Not specifiedCh5Ch11

A deep learning network was trained to automatically segment the central retinal vessel trunk and its branches (CRVT&B) from B-scans of optical coherence tomography (OCT) volume of the optic nerve head for glaucoma diagnosis.

MethodologyCohortRetrospective, deep-learning approach diagnosis studyn=Not specified for training setCh4Ch5

The segmentation network achieved a Dice coefficient of 0.81 ± 0.07 on a test set for segmenting retinal blood vessels from OCT scans.

MethodologyCohortRetrospective, deep-learning approach diagnosis studyn=Not specified for test setCh4Ch5

Is this article assigned to the wrong chapter(s)? Let us know.