Glaucomatous changes in lamina pores shape within the lamina cribrosa using wide bandwidth, femtosecond mode-locked laser OCT

Shoji, T.; Kuroda, Hiroto; Suzuki, Masayuki; Ibuki, Hisashi; Araie, Makoto; Yoneya, Shin

doi:10.1371/journal.pone.0181675

Cited by 12 publications

(10 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…RNFL thickness (AUC=0.80 ± 0.03). This result may not be surprising as the ONH do not only exhibit neural tissue changes, but also connective tissue changes, such as bending of the peripapillary sclera [5], and changes in LC morphology and pore shape/pathway [23,24]. PointNet has the advantage of capturing some of these features while minimizing the total amount of information required to establish a diagnosis.…”

Section: Discussionmentioning

confidence: 99%

Medical Application of Geometric Deep Learning for the Diagnosis of Glaucoma

Thiéry¹,

Braeu²,

Tun³

et al. 2022

Preprint

View full text Add to dashboard Cite

Purpose: (1) To assess the performance of geometric deep learning (PointNet) in diagnosing glaucoma from a single optical coherence tomography (OCT) 3D scan of the optic nerve head (ONH); (2) To compare its performance to that obtained with a standard 3D convolutional neural network (CNN), and with a gold-standard glaucoma parameter, i.e. retinal nerve ber layer (RNFL) thickness.Methods: 3D raster scans of the ONH were acquired with Spectralis OCT (Heidelberg Engineering, Heidelberg, Germany) for 477 glaucoma and 2, 296 non-glaucoma subjects at the Singapore National Eye Centre. All volumes were automatically segmented using deep learning to identify 7 major neural and connective tissues including the RNFL, the prelamina, and the lamina cribrosa (LC). Each ONH was then represented as a 3D point cloud with 1, 000 points chosen randomly from all tissue boundaries. To simplify the problem, all ONH point clouds were aligned with respect to the plane and center of Bruch's membrane opening. Geometric deep learning (PointNet) was then used to provide a glaucoma diagnosis from a single OCT point cloud. The performance of our approach (reported using the area under the receiver operating characteristic curve or AUC) was compared to that obtained with a 3D CNN, and with RNFL thickness.Results: PointNet was able to provide a robust glaucoma diagnosis solely from the ONH represented as a 3D point cloud (AUC=0.95 ± 0.01). The performance of PointNet was superior to that obtained with a standard 3D CNN (AUC=0.87 ± 0.02) and with that obtained from RNFL thickness alone (AUC=0.80 ± 0.03).Discussion: We provide a proof-of-principle for the application of geometric deep learning in the eld of glaucoma. Our technique requires signi cantly less information as input to perform better than a 3D CNN, and with an AUC superior to that obtained from RNFL thickness alone. Geometric deep learning may have wide applicability in the eld of Ophthalmology.

show abstract

Section: Discussionmentioning

confidence: 99%

Medical Application of Geometric Deep Learning for the Diagnosis of Glaucoma

Thiéry¹,

Braeu²,

Tun³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…One of the most important parts of this study was its demonstration of a method for 3D evaluation of laminar pore structure. All previously published reports have used 2D evaluation of laminar pore area with AO-SLO, [ 19 ] auto-segmentation of the laminar pores with SS-OCT [ 20 , 21 ], or have used new types of OCT. [ 22 ] However, when the disc angle is tilted, accurate evaluation of pore structure in 2D is very difficult. We found that true sphericity and the principal value (i.e., P1, 2, and 3) of the 3D structural data were significantly different in the control and glaucoma subjects in this study.…”

Section: Discussionmentioning

confidence: 99%

“…Another new technology, swept-source OCT (SS-OCT), which uses a high-penetration laser to visualize and observe the deep structure of the ONH in detail, has also been used to study the LC [ 20 , 21 ]. New OCT technologies have demonstrated that in POAG eyes, the laminar pores become significantly elongated at the anterior surface of the LC, i.e., at a depth of 40 μm, but not at a depth of 80 μm [ 22 ]. This demonstrated that SS-OCT could be used to clearly show the structure of the laminar pores in living human subjects, and could improve our understanding of the details of glaucoma pathogenesis.…”

Section: Introductionmentioning

confidence: 99%

Pilot study for three-dimensional assessment of laminar pore structure in patients with glaucoma, as measured with swept source optical coherence tomography

Omodaka

Maekawa

et al. 2018

PLoS ONE

View full text Add to dashboard Cite

PurposeTo develop a method to quantify, based on swept-source optical coherence tomography (OCT), the 3D structure of the laminar pores in patients with glaucoma.MethodsThis retrospective study examined 160 laminar pores from 8 eyes of 8 cases: 4 normal subjects and 4 open-angle glaucoma (OAG) patients. We reconstructed 3D volume data for a 3 x 3 mm disc, using a method similar to OCT angiography, and segmented the structure of the lamina cribrosa. Then, we manually segmented each laminar pore in sequential C-scan images (>90 slices at 2.6-micron intervals) with VCAT5 (RIKEN, Japan). We compared the control and OAG subjects with the Mann-Whitney U test. Differences were considered significant at p < 0.05.ResultsWe found that the laminar pores of the OAG patients had a significantly smaller average cross-sectional area, smaller 3D volume (adjusted to the average thickness of the lamina cribrosa), and higher true sphericity, and lower principal value (P1, 2, 3) of the 3D structure data (all: p < 0.0001). The topographic distribution of damaged laminar pores was consistent with the damaged area of the macular map.ConclusionWe successfully developed a method to quantify the 3D structure of the laminar pores; providing a useful tool to assess lamina cribrosa-associated risk factors for glaucoma. These findings promise to benefit future investigations into the pathomechanisms of glaucoma.

show abstract

“…Based on images obtained by swept-source optical coherence tomography (OCT), Omodaka et al found that glaucoma patients had a smaller area and 3D volume of LC pores, probably due to the expanding area of connective tissue [ 13 ]. Elongation of lamina pores was more evident in primary open-angle glaucoma eyes than in normal eyes [ 14 , 15 ].…”

Section: Introductionmentioning

confidence: 99%

Morphological Changes of Glial Lamina Cribrosa of Rats Suffering from Chronic High Intraocular Pressure

Zhang

Liu

et al. 2022

Bioengineering

View full text Add to dashboard Cite

Deformations or remodeling of the lamina cribrosa (LC) induced by elevated intraocular pressure (IOP) are associated with optic nerve injury. The quantitative analysis of the morphology changes of the LC will provide the basis for the study of the pathogenesis of glaucoma. After the chronic high-IOP rat model was induced by cauterizing episcleral veins with 5-Fluorouracil subconjunctival injection, the optic nerve head (ONH) cross sections were immunohistochemically stained at 2 w, 4 w, 8 w, and 12 w. Then the sections were imaged by a confocal microscope, and six morphological parameters of the ONH were calculated after the images were processed using Matlab. The results showed that the morphology of the ONH changed with the duration of chronic high IOP. The glial LC pore area fraction, the ratio of glial LC pore area to the glial LC tissue area, first decreased at 2 w and 4 w and then increased to the same level as the control group at 8 w and continued to increase until 12 w. The number and density of nuclei increased significantly at 8 w in the glial LC region. The results might mean the fraction of glial LC beam increased and astrocytes proliferated at the early stage of high IOP. Combined with the images of the ONH, the results showed the glial LC was damaged with the duration of chronic elevated IOP.

show abstract

Glaucomatous changes in lamina pores shape within the lamina cribrosa using wide bandwidth, femtosecond mode-locked laser OCT

Cited by 12 publications

References 39 publications

Medical Application of Geometric Deep Learning for the Diagnosis of Glaucoma

Medical Application of Geometric Deep Learning for the Diagnosis of Glaucoma

Pilot study for three-dimensional assessment of laminar pore structure in patients with glaucoma, as measured with swept source optical coherence tomography

Morphological Changes of Glial Lamina Cribrosa of Rats Suffering from Chronic High Intraocular Pressure

Contact Info

Product

Resources

About