Unsupervised Gaussian mixture-model with expectation maximization for detecting glaucomatous progression in standard automated perimetry visual fields

Siamak Yousefi, Madhusudhanan Balasubramanian, Michael H. Goldbaum, Felipe A. Medeiros, Linda M. Zangwill, Robert N. Weinreb, Jeffrey M. Liebmann, Christopher A. Girkin, Christopher Bowd

Research output: Contribution to journalArticle

15 Citations (Scopus)

Abstract

Purpose. To validate Gaussian mixture-model with expectation maximization (GEM) and variational Bayesian independent component analysis mixture-models (VIM) for detecting glaucomatous progression along visual field (VF) defect patterns (GEM– progression of patterns (POP) and VIM-POP). To compare GEM-POP and VIM-POP with other methods. Methods. GEM and VIM models separated cross-sectional abnormal VFs from 859 eyes and normal VFs from 1117 eyes into abnormal and normal clusters. Clusters were decomposed into independent axes. The confidence limit (CL) of stability was established for each axis with a set of 84 stable eyes. Sensitivity for detecting progression was assessed in a sample of 83 eyes with known progressive glaucomatous optic neuropathy (PGON). Eyes were classified as progressed if any defect pattern progressed beyond the CL of stability. Performance of GEM-POP and VIM-POP was compared to point-wise linear regression (PLR), permutation analysis of PLR (PoPLR), and linear regression (LR) of mean deviation (MD), and visual field index (VFI). Results. Sensitivity and specificity for detecting glaucomatous VFs were 89.9% and 93.8%, respectively, for GEM and 93.0% and 97.0%, respectively, for VIM. Receiver operating characteristic (ROC) curve areas for classifying progressed eyes were 0.82 for VIM-POP, 0.86 for GEM-POP, 0.81 for PoPLR, 0.69 for LR of MD, and 0.76 for LR of VFI. Conclusions. GEM-POP was significantly more sensitive to PGON than PoPLR and linear regression of MD and VFI in our sample, while providing localized progression information. Translational Relevance. Detection of glaucomatous progression can be improved by assessing longitudinal changes in localized patterns of glaucomatous defect identified by unsupervised machine learning.

Original languageEnglish (US)
Article number2
JournalTranslational Vision Science and Technology
Volume5
Issue number3
DOIs
StatePublished - May 1 2016

Fingerprint

Air cushion vehicles
Visual Field Tests
Visual Fields
Linear Models
Linear regression
Optic Nerve Diseases
Defects
Optics
ROC Curve
Independent component analysis
Regression Analysis
Regression analysis
Learning systems
Sensitivity and Specificity

All Science Journal Classification (ASJC) codes

  • Biomedical Engineering
  • Ophthalmology

Cite this

Unsupervised Gaussian mixture-model with expectation maximization for detecting glaucomatous progression in standard automated perimetry visual fields. / Yousefi, Siamak; Balasubramanian, Madhusudhanan; Goldbaum, Michael H.; Medeiros, Felipe A.; Zangwill, Linda M.; Weinreb, Robert N.; Liebmann, Jeffrey M.; Girkin, Christopher A.; Bowd, Christopher.

In: Translational Vision Science and Technology, Vol. 5, No. 3, 2, 01.05.2016.

Research output: Contribution to journalArticle

Yousefi, Siamak ; Balasubramanian, Madhusudhanan ; Goldbaum, Michael H. ; Medeiros, Felipe A. ; Zangwill, Linda M. ; Weinreb, Robert N. ; Liebmann, Jeffrey M. ; Girkin, Christopher A. ; Bowd, Christopher. / Unsupervised Gaussian mixture-model with expectation maximization for detecting glaucomatous progression in standard automated perimetry visual fields. In: Translational Vision Science and Technology. 2016 ; Vol. 5, No. 3.
@article{b51b868c977644b794f805d1afd7b61f,
title = "Unsupervised Gaussian mixture-model with expectation maximization for detecting glaucomatous progression in standard automated perimetry visual fields",
abstract = "Purpose. To validate Gaussian mixture-model with expectation maximization (GEM) and variational Bayesian independent component analysis mixture-models (VIM) for detecting glaucomatous progression along visual field (VF) defect patterns (GEM– progression of patterns (POP) and VIM-POP). To compare GEM-POP and VIM-POP with other methods. Methods. GEM and VIM models separated cross-sectional abnormal VFs from 859 eyes and normal VFs from 1117 eyes into abnormal and normal clusters. Clusters were decomposed into independent axes. The confidence limit (CL) of stability was established for each axis with a set of 84 stable eyes. Sensitivity for detecting progression was assessed in a sample of 83 eyes with known progressive glaucomatous optic neuropathy (PGON). Eyes were classified as progressed if any defect pattern progressed beyond the CL of stability. Performance of GEM-POP and VIM-POP was compared to point-wise linear regression (PLR), permutation analysis of PLR (PoPLR), and linear regression (LR) of mean deviation (MD), and visual field index (VFI). Results. Sensitivity and specificity for detecting glaucomatous VFs were 89.9{\%} and 93.8{\%}, respectively, for GEM and 93.0{\%} and 97.0{\%}, respectively, for VIM. Receiver operating characteristic (ROC) curve areas for classifying progressed eyes were 0.82 for VIM-POP, 0.86 for GEM-POP, 0.81 for PoPLR, 0.69 for LR of MD, and 0.76 for LR of VFI. Conclusions. GEM-POP was significantly more sensitive to PGON than PoPLR and linear regression of MD and VFI in our sample, while providing localized progression information. Translational Relevance. Detection of glaucomatous progression can be improved by assessing longitudinal changes in localized patterns of glaucomatous defect identified by unsupervised machine learning.",
author = "Siamak Yousefi and Madhusudhanan Balasubramanian and Goldbaum, {Michael H.} and Medeiros, {Felipe A.} and Zangwill, {Linda M.} and Weinreb, {Robert N.} and Liebmann, {Jeffrey M.} and Girkin, {Christopher A.} and Christopher Bowd",
year = "2016",
month = "5",
day = "1",
doi = "10.1167/tvst.5.3.2",
language = "English (US)",
volume = "5",
journal = "Translational Vision Science and Technology",
issn = "2164-2591",
publisher = "Association for Research in Vision and Ophthalmology Inc.",
number = "3",

}

TY - JOUR

T1 - Unsupervised Gaussian mixture-model with expectation maximization for detecting glaucomatous progression in standard automated perimetry visual fields

AU - Yousefi, Siamak

AU - Balasubramanian, Madhusudhanan

AU - Goldbaum, Michael H.

AU - Medeiros, Felipe A.

AU - Zangwill, Linda M.

AU - Weinreb, Robert N.

AU - Liebmann, Jeffrey M.

AU - Girkin, Christopher A.

AU - Bowd, Christopher

PY - 2016/5/1

Y1 - 2016/5/1

N2 - Purpose. To validate Gaussian mixture-model with expectation maximization (GEM) and variational Bayesian independent component analysis mixture-models (VIM) for detecting glaucomatous progression along visual field (VF) defect patterns (GEM– progression of patterns (POP) and VIM-POP). To compare GEM-POP and VIM-POP with other methods. Methods. GEM and VIM models separated cross-sectional abnormal VFs from 859 eyes and normal VFs from 1117 eyes into abnormal and normal clusters. Clusters were decomposed into independent axes. The confidence limit (CL) of stability was established for each axis with a set of 84 stable eyes. Sensitivity for detecting progression was assessed in a sample of 83 eyes with known progressive glaucomatous optic neuropathy (PGON). Eyes were classified as progressed if any defect pattern progressed beyond the CL of stability. Performance of GEM-POP and VIM-POP was compared to point-wise linear regression (PLR), permutation analysis of PLR (PoPLR), and linear regression (LR) of mean deviation (MD), and visual field index (VFI). Results. Sensitivity and specificity for detecting glaucomatous VFs were 89.9% and 93.8%, respectively, for GEM and 93.0% and 97.0%, respectively, for VIM. Receiver operating characteristic (ROC) curve areas for classifying progressed eyes were 0.82 for VIM-POP, 0.86 for GEM-POP, 0.81 for PoPLR, 0.69 for LR of MD, and 0.76 for LR of VFI. Conclusions. GEM-POP was significantly more sensitive to PGON than PoPLR and linear regression of MD and VFI in our sample, while providing localized progression information. Translational Relevance. Detection of glaucomatous progression can be improved by assessing longitudinal changes in localized patterns of glaucomatous defect identified by unsupervised machine learning.

AB - Purpose. To validate Gaussian mixture-model with expectation maximization (GEM) and variational Bayesian independent component analysis mixture-models (VIM) for detecting glaucomatous progression along visual field (VF) defect patterns (GEM– progression of patterns (POP) and VIM-POP). To compare GEM-POP and VIM-POP with other methods. Methods. GEM and VIM models separated cross-sectional abnormal VFs from 859 eyes and normal VFs from 1117 eyes into abnormal and normal clusters. Clusters were decomposed into independent axes. The confidence limit (CL) of stability was established for each axis with a set of 84 stable eyes. Sensitivity for detecting progression was assessed in a sample of 83 eyes with known progressive glaucomatous optic neuropathy (PGON). Eyes were classified as progressed if any defect pattern progressed beyond the CL of stability. Performance of GEM-POP and VIM-POP was compared to point-wise linear regression (PLR), permutation analysis of PLR (PoPLR), and linear regression (LR) of mean deviation (MD), and visual field index (VFI). Results. Sensitivity and specificity for detecting glaucomatous VFs were 89.9% and 93.8%, respectively, for GEM and 93.0% and 97.0%, respectively, for VIM. Receiver operating characteristic (ROC) curve areas for classifying progressed eyes were 0.82 for VIM-POP, 0.86 for GEM-POP, 0.81 for PoPLR, 0.69 for LR of MD, and 0.76 for LR of VFI. Conclusions. GEM-POP was significantly more sensitive to PGON than PoPLR and linear regression of MD and VFI in our sample, while providing localized progression information. Translational Relevance. Detection of glaucomatous progression can be improved by assessing longitudinal changes in localized patterns of glaucomatous defect identified by unsupervised machine learning.

UR - http://www.scopus.com/inward/record.url?scp=85012144233&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85012144233&partnerID=8YFLogxK

U2 - 10.1167/tvst.5.3.2

DO - 10.1167/tvst.5.3.2

M3 - Article

VL - 5

JO - Translational Vision Science and Technology

JF - Translational Vision Science and Technology

SN - 2164-2591

IS - 3

M1 - 2

ER -