Learning from data

Recognizing glaucomatous defect patterns and detecting progression from visual field measurements

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

Research output: Contribution to journalArticle

17 Citations (Scopus)

Abstract

A hierarchical approach to learn from visual field data was adopted to identify glaucomatous visual field defect patterns and to detect glaucomatous progression. The analysis pipeline included three stages, namely, clustering, glaucoma boundary limit detection, and glaucoma progression detection testing. First, cross-sectional visual field tests collected from each subject were clustered using a mixture of Gaussians and model parameters were estimated using expectation maximization. The visual field clusters were further estimated to recognize glaucomatous visual field defect patterns by decomposing each cluster into several axes. The glaucoma visual field defect patterns along each axis then were identified. To derive a definition of progression, the longitudinal visual fields of stable glaucoma eyes on the abnormal cluster axes were projected and the slope was approximated using linear regression (LR) to determine the confidence limit of each axis. For glaucoma progression detection, the longitudinal visual fields of each eye on the abnormal cluster axes were projected and the slope was approximated by LR. Progression was assigned if the progression rate was greater than the boundary limit of the stable eyes; otherwise, stability was assumed. The proposed method was compared to a recently developed progression detection method and to clinically available glaucoma progression detection software. The clinical accuracy of the proposed pipeline was as good as or better than the currently available methods.

Original languageEnglish (US)
Article number6781613
Pages (from-to)2112-2124
Number of pages13
JournalIEEE Transactions on Biomedical Engineering
Volume61
Issue number7
DOIs
StatePublished - Jan 1 2014

Fingerprint

Linear regression
Defects
Pipelines
Testing

All Science Journal Classification (ASJC) codes

  • Biomedical Engineering

Cite this

Yousefi, S., Goldbaum, M. H., Balasubramanian, M., Medeiros, F. A., Zangwill, L. M., Liebmann, J. M., ... Bowd, C. (2014). Learning from data: Recognizing glaucomatous defect patterns and detecting progression from visual field measurements. IEEE Transactions on Biomedical Engineering, 61(7), 2112-2124. [6781613]. https://doi.org/10.1109/TBME.2014.2314714

Learning from data : Recognizing glaucomatous defect patterns and detecting progression from visual field measurements. / Yousefi, Siamak; Goldbaum, Michael H.; Balasubramanian, Madhusudhanan; Medeiros, Felipe A.; Zangwill, Linda M.; Liebmann, Jeffrey M.; Girkin, Christopher A.; Weinreb, Robert N.; Bowd, Christopher.

In: IEEE Transactions on Biomedical Engineering, Vol. 61, No. 7, 6781613, 01.01.2014, p. 2112-2124.

Research output: Contribution to journalArticle

Yousefi, S, Goldbaum, MH, Balasubramanian, M, Medeiros, FA, Zangwill, LM, Liebmann, JM, Girkin, CA, Weinreb, RN & Bowd, C 2014, 'Learning from data: Recognizing glaucomatous defect patterns and detecting progression from visual field measurements', IEEE Transactions on Biomedical Engineering, vol. 61, no. 7, 6781613, pp. 2112-2124. https://doi.org/10.1109/TBME.2014.2314714
Yousefi, Siamak ; Goldbaum, Michael H. ; Balasubramanian, Madhusudhanan ; Medeiros, Felipe A. ; Zangwill, Linda M. ; Liebmann, Jeffrey M. ; Girkin, Christopher A. ; Weinreb, Robert N. ; Bowd, Christopher. / Learning from data : Recognizing glaucomatous defect patterns and detecting progression from visual field measurements. In: IEEE Transactions on Biomedical Engineering. 2014 ; Vol. 61, No. 7. pp. 2112-2124.
@article{a6d2c64f2d604b9da5e87c2e333875c5,
title = "Learning from data: Recognizing glaucomatous defect patterns and detecting progression from visual field measurements",
abstract = "A hierarchical approach to learn from visual field data was adopted to identify glaucomatous visual field defect patterns and to detect glaucomatous progression. The analysis pipeline included three stages, namely, clustering, glaucoma boundary limit detection, and glaucoma progression detection testing. First, cross-sectional visual field tests collected from each subject were clustered using a mixture of Gaussians and model parameters were estimated using expectation maximization. The visual field clusters were further estimated to recognize glaucomatous visual field defect patterns by decomposing each cluster into several axes. The glaucoma visual field defect patterns along each axis then were identified. To derive a definition of progression, the longitudinal visual fields of stable glaucoma eyes on the abnormal cluster axes were projected and the slope was approximated using linear regression (LR) to determine the confidence limit of each axis. For glaucoma progression detection, the longitudinal visual fields of each eye on the abnormal cluster axes were projected and the slope was approximated by LR. Progression was assigned if the progression rate was greater than the boundary limit of the stable eyes; otherwise, stability was assumed. The proposed method was compared to a recently developed progression detection method and to clinically available glaucoma progression detection software. The clinical accuracy of the proposed pipeline was as good as or better than the currently available methods.",
author = "Siamak Yousefi and Goldbaum, {Michael H.} and Madhusudhanan Balasubramanian and Medeiros, {Felipe A.} and Zangwill, {Linda M.} and Liebmann, {Jeffrey M.} and Girkin, {Christopher A.} and Weinreb, {Robert N.} and Christopher Bowd",
year = "2014",
month = "1",
day = "1",
doi = "10.1109/TBME.2014.2314714",
language = "English (US)",
volume = "61",
pages = "2112--2124",
journal = "IEEE Transactions on Biomedical Engineering",
issn = "0018-9294",
publisher = "IEEE Computer Society",
number = "7",

}

TY - JOUR

T1 - Learning from data

T2 - Recognizing glaucomatous defect patterns and detecting progression from visual field measurements

AU - Yousefi, Siamak

AU - Goldbaum, Michael H.

AU - Balasubramanian, Madhusudhanan

AU - Medeiros, Felipe A.

AU - Zangwill, Linda M.

AU - Liebmann, Jeffrey M.

AU - Girkin, Christopher A.

AU - Weinreb, Robert N.

AU - Bowd, Christopher

PY - 2014/1/1

Y1 - 2014/1/1

N2 - A hierarchical approach to learn from visual field data was adopted to identify glaucomatous visual field defect patterns and to detect glaucomatous progression. The analysis pipeline included three stages, namely, clustering, glaucoma boundary limit detection, and glaucoma progression detection testing. First, cross-sectional visual field tests collected from each subject were clustered using a mixture of Gaussians and model parameters were estimated using expectation maximization. The visual field clusters were further estimated to recognize glaucomatous visual field defect patterns by decomposing each cluster into several axes. The glaucoma visual field defect patterns along each axis then were identified. To derive a definition of progression, the longitudinal visual fields of stable glaucoma eyes on the abnormal cluster axes were projected and the slope was approximated using linear regression (LR) to determine the confidence limit of each axis. For glaucoma progression detection, the longitudinal visual fields of each eye on the abnormal cluster axes were projected and the slope was approximated by LR. Progression was assigned if the progression rate was greater than the boundary limit of the stable eyes; otherwise, stability was assumed. The proposed method was compared to a recently developed progression detection method and to clinically available glaucoma progression detection software. The clinical accuracy of the proposed pipeline was as good as or better than the currently available methods.

AB - A hierarchical approach to learn from visual field data was adopted to identify glaucomatous visual field defect patterns and to detect glaucomatous progression. The analysis pipeline included three stages, namely, clustering, glaucoma boundary limit detection, and glaucoma progression detection testing. First, cross-sectional visual field tests collected from each subject were clustered using a mixture of Gaussians and model parameters were estimated using expectation maximization. The visual field clusters were further estimated to recognize glaucomatous visual field defect patterns by decomposing each cluster into several axes. The glaucoma visual field defect patterns along each axis then were identified. To derive a definition of progression, the longitudinal visual fields of stable glaucoma eyes on the abnormal cluster axes were projected and the slope was approximated using linear regression (LR) to determine the confidence limit of each axis. For glaucoma progression detection, the longitudinal visual fields of each eye on the abnormal cluster axes were projected and the slope was approximated by LR. Progression was assigned if the progression rate was greater than the boundary limit of the stable eyes; otherwise, stability was assumed. The proposed method was compared to a recently developed progression detection method and to clinically available glaucoma progression detection software. The clinical accuracy of the proposed pipeline was as good as or better than the currently available methods.

UR - http://www.scopus.com/inward/record.url?scp=84903157123&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84903157123&partnerID=8YFLogxK

U2 - 10.1109/TBME.2014.2314714

DO - 10.1109/TBME.2014.2314714

M3 - Article

VL - 61

SP - 2112

EP - 2124

JO - IEEE Transactions on Biomedical Engineering

JF - IEEE Transactions on Biomedical Engineering

SN - 0018-9294

IS - 7

M1 - 6781613

ER -