Literature DB >> 35202616

Policy-Driven, Multimodal Deep Learning for Predicting Visual Fields from the Optic Disc and OCT Imaging.

Yuka Kihara1, Giovanni Montesano2, Andrew Chen1, Nishani Amerasinghe3, Chrysostomos Dimitriou4, Aby Jacob3, Almira Chabi5, David P Crabb6, Aaron Y Lee7.   

Abstract

PURPOSE: To develop and validate a deep learning (DL) system for predicting each point on visual fields (VFs) from disc and OCT imaging and derive a structure-function mapping.
DESIGN: Retrospective, cross-sectional database study. PARTICIPANTS: A total of 6437 patients undergoing routine care for glaucoma in 3 clinical sites in the United Kingdom.
METHODS: OCT and infrared reflectance (IR) optic disc imaging were paired with the closest VF within 7 days. EfficientNet B2 was used to train 2 single-modality DL models to predict each of the 52 sensitivity points on the 24-2 VF pattern. A policy DL model was designed and trained to fuse the 2 model predictions. MAIN OUTCOME MEASURES: Pointwise mean absolute error (PMAE).
RESULTS: A total of 5078 imaging scans to VF pairs were used as a held-out test set to measure the final performance. The improvement in PMAE with the policy model was 0.485 (0.438, 0.533) decibels (dB) compared with the IR image of the disc alone and 0.060 (0.047, 0.073) dB with to the OCT alone. The improvement with the policy fusion model was statistically significant (P < 0.0001). Occlusion masking shows that the DL models learned the correct structure-function mapping in a data-driven, feature agnostic fashion.
CONCLUSIONS: The multimodal, policy DL model performed the best; it provided explainable maps of its confidence in fusing data from single modalities and provides a pathway for probing the structure-function relationship in glaucoma.
Copyright © 2022 American Academy of Ophthalmology. Published by Elsevier Inc. All rights reserved.

Entities:  

Keywords:  Artificial intelligence; Deep learning; Glaucoma; OCT; Perimetry; Structure–function; Visual field

Mesh:

Year:  2022        PMID: 35202616      PMCID: PMC9233104          DOI: 10.1016/j.ophtha.2022.02.017

Source DB:  PubMed          Journal:  Ophthalmology        ISSN: 0161-6420            Impact factor:   14.277


  39 in total

1.  An anatomically customizable computational model relating the visual field to the optic nerve head in individual eyes.

Authors:  Jonathan Denniss; Allison M McKendrick; Andrew Turpin
Journal:  Invest Ophthalmol Vis Sci       Date:  2012-10-09       Impact factor: 4.799

2.  The visualFields package: a tool for analysis and visualization of visual fields.

Authors:  Iván Marín-Franch; William H Swanson
Journal:  J Vis       Date:  2013-03-14       Impact factor: 2.240

3.  Mapping the visual field to the optic disc in normal tension glaucoma eyes.

Authors:  D F Garway-Heath; D Poinoosawmy; F W Fitzke; R A Hitchings
Journal:  Ophthalmology       Date:  2000-10       Impact factor: 12.079

4.  The onset and evolution of glaucomatous visual field defects.

Authors:  W M Hart; B Becker
Journal:  Ophthalmology       Date:  1982-03       Impact factor: 12.079

5.  Combining ganglion cell topology and data of patients with glaucoma to determine a structure-function map.

Authors:  Andrew Turpin; Geoff P Sampson; Allison M McKendrick
Journal:  Invest Ophthalmol Vis Sci       Date:  2009-03-25       Impact factor: 4.799

6.  Estimating Global Visual Field Indices in Glaucoma by Combining Macula and Optic Disc OCT Scans Using 3-Dimensional Convolutional Neural Networks.

Authors:  Hsin-Hao Yu; Stefan R Maetschke; Bhavna J Antony; Hiroshi Ishikawa; Gadi Wollstein; Joel S Schuman; Rahil Garnavi
Journal:  Ophthalmol Glaucoma       Date:  2020-07-11

7.  Visual Field Inference From Optical Coherence Tomography Using Deep Learning Algorithms: A Comparison Between Devices.

Authors:  Jonghoon Shin; Sungjoon Kim; Jinmi Kim; Keunheung Park
Journal:  Transl Vis Sci Technol       Date:  2021-06-01       Impact factor: 3.283

8.  A deep learning approach to predict visual field using optical coherence tomography.

Authors:  Keunheung Park; Jinmi Kim; Jiwoong Lee
Journal:  PLoS One       Date:  2020-07-06       Impact factor: 3.240

9.  Estimating visual field loss from monoscopic optic disc photography using deep learning model.

Authors:  Jinho Lee; Yong Woo Kim; Ahnul Ha; Young Kook Kim; Ki Ho Park; Hyuk Jin Choi; Jin Wook Jeoung
Journal:  Sci Rep       Date:  2020-12-03       Impact factor: 4.379

10.  Revisiting the Drasdo Model: Implications for Structure-Function Analysis of the Macular Region.

Authors:  Giovanni Montesano; Giovanni Ometto; Ruth E Hogg; Luca M Rossetti; David F Garway-Heath; David P Crabb
Journal:  Transl Vis Sci Technol       Date:  2020-09-14       Impact factor: 3.283

View more

北京卡尤迪生物科技股份有限公司 © 2022-2023.