An Artificial Intelligence Approach to Detect Visual Field Progression in Glaucoma Based on Spatial Pattern Analysis

Mengyu Wang, Lucy Q. Shen, Louis R. Pasquale, Paul Petrakos, Sydney Formica, Michael Boland, Sarah R. Wellik, Carlos Gustavo De Moraes, Jonathan S. Myers, Osamah Saeedi, Hui Wang, Neda Baniasadi, Dian Li, Jorryt Tichelaar, Peter J. Bex, Tobias Elze

Research output: Contribution to journalArticle

Abstract

Purpose: To detect visual field (VF) progression by analyzing spatial pattern changes. Methods: We selected 12,217 eyes from 7360 patients with at least five reliable 24-2 VFs and 5 years of follow-up with an interval of at least 6 months. VFs were decomposed into 16 archetype patterns previously derived by artificial intelligence techniques. Linear regressions were applied to the 16 archetype weights of VF series over time. We defined progression as the decrease rate of the normal archetype or any increase rate of the 15 VF defect archetypes to be outside normal limits. The archetype method was compared with mean deviation (MD) slope, Advanced Glaucoma Intervention Study (AGIS) scoring, Collaborative Initial Glaucoma Treatment Study (CIGTS) scoring, and the permutation of pointwise linear regression (PoPLR), and was validated by a subset of VFs assessed by three glaucoma specialists. Results: In the method development cohort of 11,817 eyes, the archetype method agreed more with MD slope (kappa: 0.37) and PoPLR (0.33) than AGIS (0.12) and CIGTS (0.22). The most frequently progressed patterns included decreased normal pattern (63.7%), and increased nasal steps (16.4%), altitudinal loss (15.9%), superior-peripheral defect (12.1%), paracentral/central defects (10.5%), and near total loss (10.4%). In the clinical validation cohort of 397 eyes with 27.5% of confirmed progression, the agreement (kappa) and accuracy (mean of hit rate and correct rejection rate) of the archetype method (0.51 and 0.77) significantly (P < 0.001 for all) outperformed AGIS (0.06 and 0.52), CIGTS (0.24 and 0.59), MD slope (0.21 and 0.59), and PoPLR (0.26 and 0.60). Conclusions: The archetype method can inform clinicians of VF progression patterns.

Original languageEnglish (US)
Pages (from-to)365-375
Number of pages11
JournalInvestigative ophthalmology & visual science
Volume60
Issue number1
DOIs
StatePublished - Jan 2 2019

Fingerprint

Spatial Analysis
Artificial Intelligence
Visual Fields
Glaucoma
Linear Models
Nose
Weights and Measures
Therapeutics

ASJC Scopus subject areas

  • Ophthalmology
  • Sensory Systems
  • Cellular and Molecular Neuroscience

Cite this

An Artificial Intelligence Approach to Detect Visual Field Progression in Glaucoma Based on Spatial Pattern Analysis. / Wang, Mengyu; Shen, Lucy Q.; Pasquale, Louis R.; Petrakos, Paul; Formica, Sydney; Boland, Michael; Wellik, Sarah R.; De Moraes, Carlos Gustavo; Myers, Jonathan S.; Saeedi, Osamah; Wang, Hui; Baniasadi, Neda; Li, Dian; Tichelaar, Jorryt; Bex, Peter J.; Elze, Tobias.

In: Investigative ophthalmology & visual science, Vol. 60, No. 1, 02.01.2019, p. 365-375.

Research output: Contribution to journalArticle

Wang, M, Shen, LQ, Pasquale, LR, Petrakos, P, Formica, S, Boland, M, Wellik, SR, De Moraes, CG, Myers, JS, Saeedi, O, Wang, H, Baniasadi, N, Li, D, Tichelaar, J, Bex, PJ & Elze, T 2019, 'An Artificial Intelligence Approach to Detect Visual Field Progression in Glaucoma Based on Spatial Pattern Analysis', Investigative ophthalmology & visual science, vol. 60, no. 1, pp. 365-375. https://doi.org/10.1167/iovs.18-25568
Wang, Mengyu ; Shen, Lucy Q. ; Pasquale, Louis R. ; Petrakos, Paul ; Formica, Sydney ; Boland, Michael ; Wellik, Sarah R. ; De Moraes, Carlos Gustavo ; Myers, Jonathan S. ; Saeedi, Osamah ; Wang, Hui ; Baniasadi, Neda ; Li, Dian ; Tichelaar, Jorryt ; Bex, Peter J. ; Elze, Tobias. / An Artificial Intelligence Approach to Detect Visual Field Progression in Glaucoma Based on Spatial Pattern Analysis. In: Investigative ophthalmology & visual science. 2019 ; Vol. 60, No. 1. pp. 365-375.
@article{91c2f476647e43cf9673d9324fecee08,
title = "An Artificial Intelligence Approach to Detect Visual Field Progression in Glaucoma Based on Spatial Pattern Analysis",
abstract = "Purpose: To detect visual field (VF) progression by analyzing spatial pattern changes. Methods: We selected 12,217 eyes from 7360 patients with at least five reliable 24-2 VFs and 5 years of follow-up with an interval of at least 6 months. VFs were decomposed into 16 archetype patterns previously derived by artificial intelligence techniques. Linear regressions were applied to the 16 archetype weights of VF series over time. We defined progression as the decrease rate of the normal archetype or any increase rate of the 15 VF defect archetypes to be outside normal limits. The archetype method was compared with mean deviation (MD) slope, Advanced Glaucoma Intervention Study (AGIS) scoring, Collaborative Initial Glaucoma Treatment Study (CIGTS) scoring, and the permutation of pointwise linear regression (PoPLR), and was validated by a subset of VFs assessed by three glaucoma specialists. Results: In the method development cohort of 11,817 eyes, the archetype method agreed more with MD slope (kappa: 0.37) and PoPLR (0.33) than AGIS (0.12) and CIGTS (0.22). The most frequently progressed patterns included decreased normal pattern (63.7{\%}), and increased nasal steps (16.4{\%}), altitudinal loss (15.9{\%}), superior-peripheral defect (12.1{\%}), paracentral/central defects (10.5{\%}), and near total loss (10.4{\%}). In the clinical validation cohort of 397 eyes with 27.5{\%} of confirmed progression, the agreement (kappa) and accuracy (mean of hit rate and correct rejection rate) of the archetype method (0.51 and 0.77) significantly (P < 0.001 for all) outperformed AGIS (0.06 and 0.52), CIGTS (0.24 and 0.59), MD slope (0.21 and 0.59), and PoPLR (0.26 and 0.60). Conclusions: The archetype method can inform clinicians of VF progression patterns.",
author = "Mengyu Wang and Shen, {Lucy Q.} and Pasquale, {Louis R.} and Paul Petrakos and Sydney Formica and Michael Boland and Wellik, {Sarah R.} and {De Moraes}, {Carlos Gustavo} and Myers, {Jonathan S.} and Osamah Saeedi and Hui Wang and Neda Baniasadi and Dian Li and Jorryt Tichelaar and Bex, {Peter J.} and Tobias Elze",
year = "2019",
month = "1",
day = "2",
doi = "10.1167/iovs.18-25568",
language = "English (US)",
volume = "60",
pages = "365--375",
journal = "Investigative Ophthalmology and Visual Science",
issn = "0146-0404",
publisher = "Association for Research in Vision and Ophthalmology Inc.",
number = "1",

}

TY - JOUR

T1 - An Artificial Intelligence Approach to Detect Visual Field Progression in Glaucoma Based on Spatial Pattern Analysis

AU - Wang, Mengyu

AU - Shen, Lucy Q.

AU - Pasquale, Louis R.

AU - Petrakos, Paul

AU - Formica, Sydney

AU - Boland, Michael

AU - Wellik, Sarah R.

AU - De Moraes, Carlos Gustavo

AU - Myers, Jonathan S.

AU - Saeedi, Osamah

AU - Wang, Hui

AU - Baniasadi, Neda

AU - Li, Dian

AU - Tichelaar, Jorryt

AU - Bex, Peter J.

AU - Elze, Tobias

PY - 2019/1/2

Y1 - 2019/1/2

N2 - Purpose: To detect visual field (VF) progression by analyzing spatial pattern changes. Methods: We selected 12,217 eyes from 7360 patients with at least five reliable 24-2 VFs and 5 years of follow-up with an interval of at least 6 months. VFs were decomposed into 16 archetype patterns previously derived by artificial intelligence techniques. Linear regressions were applied to the 16 archetype weights of VF series over time. We defined progression as the decrease rate of the normal archetype or any increase rate of the 15 VF defect archetypes to be outside normal limits. The archetype method was compared with mean deviation (MD) slope, Advanced Glaucoma Intervention Study (AGIS) scoring, Collaborative Initial Glaucoma Treatment Study (CIGTS) scoring, and the permutation of pointwise linear regression (PoPLR), and was validated by a subset of VFs assessed by three glaucoma specialists. Results: In the method development cohort of 11,817 eyes, the archetype method agreed more with MD slope (kappa: 0.37) and PoPLR (0.33) than AGIS (0.12) and CIGTS (0.22). The most frequently progressed patterns included decreased normal pattern (63.7%), and increased nasal steps (16.4%), altitudinal loss (15.9%), superior-peripheral defect (12.1%), paracentral/central defects (10.5%), and near total loss (10.4%). In the clinical validation cohort of 397 eyes with 27.5% of confirmed progression, the agreement (kappa) and accuracy (mean of hit rate and correct rejection rate) of the archetype method (0.51 and 0.77) significantly (P < 0.001 for all) outperformed AGIS (0.06 and 0.52), CIGTS (0.24 and 0.59), MD slope (0.21 and 0.59), and PoPLR (0.26 and 0.60). Conclusions: The archetype method can inform clinicians of VF progression patterns.

AB - Purpose: To detect visual field (VF) progression by analyzing spatial pattern changes. Methods: We selected 12,217 eyes from 7360 patients with at least five reliable 24-2 VFs and 5 years of follow-up with an interval of at least 6 months. VFs were decomposed into 16 archetype patterns previously derived by artificial intelligence techniques. Linear regressions were applied to the 16 archetype weights of VF series over time. We defined progression as the decrease rate of the normal archetype or any increase rate of the 15 VF defect archetypes to be outside normal limits. The archetype method was compared with mean deviation (MD) slope, Advanced Glaucoma Intervention Study (AGIS) scoring, Collaborative Initial Glaucoma Treatment Study (CIGTS) scoring, and the permutation of pointwise linear regression (PoPLR), and was validated by a subset of VFs assessed by three glaucoma specialists. Results: In the method development cohort of 11,817 eyes, the archetype method agreed more with MD slope (kappa: 0.37) and PoPLR (0.33) than AGIS (0.12) and CIGTS (0.22). The most frequently progressed patterns included decreased normal pattern (63.7%), and increased nasal steps (16.4%), altitudinal loss (15.9%), superior-peripheral defect (12.1%), paracentral/central defects (10.5%), and near total loss (10.4%). In the clinical validation cohort of 397 eyes with 27.5% of confirmed progression, the agreement (kappa) and accuracy (mean of hit rate and correct rejection rate) of the archetype method (0.51 and 0.77) significantly (P < 0.001 for all) outperformed AGIS (0.06 and 0.52), CIGTS (0.24 and 0.59), MD slope (0.21 and 0.59), and PoPLR (0.26 and 0.60). Conclusions: The archetype method can inform clinicians of VF progression patterns.

UR - http://www.scopus.com/inward/record.url?scp=85060529303&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85060529303&partnerID=8YFLogxK

U2 - 10.1167/iovs.18-25568

DO - 10.1167/iovs.18-25568

M3 - Article

C2 - 30682206

AN - SCOPUS:85060529303

VL - 60

SP - 365

EP - 375

JO - Investigative Ophthalmology and Visual Science

JF - Investigative Ophthalmology and Visual Science

SN - 0146-0404

IS - 1

ER -