Medical Application of Geometric Deep Learning for the Diagnosis of Glaucoma

• URL: http://arxiv.org/abs/2204.07004v1
• Date: Thu, 14 Apr 2022 14:55:25 GMT
• Title: Medical Application of Geometric Deep Learning for the Diagnosis of Glaucoma
• Authors: Alexandre H. Thiery, Fabian Braeu, Tin A. Tun, Tin Aung, Michael J.A. Girard
• Abstract summary: 3D scans of the optic nerve head (ONH) were acquired with Spectralis OCT 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. PointNet was able to provide a robust glaucoma diagnosis solely from the ONH represented as a 3D point cloud.
• Score: 60.42955087779866
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: 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 fiber layer (RNFL) thickness. Methods: 3D raster scans of the ONH were acquired with Spectralis OCT 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 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=95%). The performance of PointNet was superior to that obtained with a standard 3D CNN (AUC=87%) and with that obtained from RNFL thickness alone (AUC=80%). Discussion: We provide a proof-of-principle for the application of geometric deep learning in the field of glaucoma. Our technique requires significantly 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 field of Ophthalmology.

Related papers

• N-BVH: Neural ray queries with bounding volume hierarchies [51.430495562430565]
  In 3D computer graphics, the bulk of a scene's memory usage is due to polygons and textures. We devise N-BVH, a neural compression architecture designed to answer arbitrary ray queries in 3D. Our method provides faithful approximations of visibility, depth, and appearance attributes.
  arXiv  Detail & Related papers  (2024-05-25T13:54:34Z)
• The 3D Structural Phenotype of the Glaucomatous Optic Nerve Head and its Relationship with The Severity of Visual Field Damage [45.923831389099696]
  We observed that the majority of ONH structural changes occurred in the early glaucoma stage, followed by a plateau effect in the later stages. Using PointNet, we also found that 3D ONH structural changes were present in both neural and connective tissues.
  arXiv  Detail & Related papers  (2023-01-07T12:28:43Z)
• Are Macula or Optic Nerve Head Structures better at Diagnosing Glaucoma? An Answer using AI and Wide-Field Optical Coherence Tomography [48.7576911714538]
  We developed a deep learning algorithm to automatically segment structures of the optic nerve head (ONH) and macula in 3D wide-field OCT scans. Our classification algorithm was able to segment ONH and macular tissues with a DC of 0.94 $pm$ 0.003. This may encourage the mainstream adoption of 3D wide-field OCT scans.
  arXiv  Detail & Related papers  (2022-10-13T01:51:29Z)
• AI-based Clinical Assessment of Optic Nerve Head Robustness Superseding Biomechanical Testing [54.306443917863355]
  We propose an AI-driven approach that can assess the robustness of a given ONH solely from a single OCT scan of the ONH. Longitudinal studies should establish whether ONH robustness could help us identify fast visual field loss progressors.
  arXiv  Detail & Related papers  (2022-06-09T11:29:28Z)
• Geometric Deep Learning to Identify the Critical 3D Structural Features of the Optic Nerve Head for Glaucoma Diagnosis [52.06403518904579]
  The optic nerve head (ONH) undergoes complex and deep 3D morphological changes during the development and progression of glaucoma. We used PointNet and dynamic graph convolutional neural network (DGCNN) to diagnose glaucoma from 3D ONH point clouds. Our approach may have strong potential to be used in clinical applications for the diagnosis and prognosis of a wide range of ophthalmic disorders.
  arXiv  Detail & Related papers  (2022-04-14T12:52:10Z)
• 3D Structural Analysis of the Optic Nerve Head to Robustly Discriminate Between Papilledema and Optic Disc Drusen [44.754910718620295]
  We developed a deep learning algorithm to identify major tissue structures of the optic nerve head (ONH) in 3D optical coherence tomography ( OCT) scans. A classification algorithm was designed using 150 OCT volumes to perform 3-class classifications (1: ODD, 2: papilledema, 3: healthy) strictly from their drusen and prelamina swelling scores. Our AI approach accurately discriminated ODD from papilledema, using a single OCT scan.
  arXiv  Detail & Related papers  (2021-12-18T17:05:53Z)
• The Three-Dimensional Structural Configuration of the Central Retinal Vessel Trunk and Branches as a Glaucoma Biomarker [41.97805846007449]
  We trained a deep learning network to automatically segment the CRVT&B from the B-scans of the optical coherence tomography volume of the optic nerve head (ONH) The 3D and 2D diagnostic networks were able to differentiate glaucoma from non-glaucoma subjects with accuracies of 82.7% and 83.3%, respectively.
  arXiv  Detail & Related papers  (2021-11-07T04:41:49Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.