Recognizing patterns of visual field loss using unsupervised machine learning

Siamak Yousefi, Michael H. Goldbaum, Linda M. Zangwill, Felipe A. Medeiros, Christopher Bowd

Research output: Chapter in Book/Report/Conference proceedingConference contribution

7 Citations (Scopus)

Abstract

Glaucoma is a potentially blinding optic neuropathy that results in a decrease in visual sensitivity. Visual field abnormalities (decreased visual sensitivity on psychophysical tests) are the primary means of glaucoma diagnosis. One form of visual field testing is Frequency Doubling Technology (FDT) that tests sensitivity at 52 points within the visual field. Like other psychophysical tests used in clinical practice, FDT results yield specific patterns of defect indicative of the disease. We used Gaussian Mixture Model with Expectation Maximization (GEM), (EM is used to estimate the model parameters) to automatically separate FDT data into clusters of normal and abnormal eyes. Principal component analysis (PCA) was used to decompose each cluster into different axes (patterns). FDT measurements were obtained from 1,190 eyes with normal FDT results and 786 eyes with abnormal (i.e., glaucomatous) FDT results, recruited from a university-based, longitudinal, multi-center, clinical study on glaucoma. The GEM input was the 52-point FDT threshold sensitivities for all eyes. The optimal GEM model separated the FDT fields into 3 clusters. Cluster 1 contained 94% normal fields (94% specificity) and clusters 2 and 3 combined, contained 77% abnormal fields (77% sensitivity). For clusters 1, 2 and 3 the optimal number of PCA-identified axes were 2, 2 and 5, respectively. GEM with PCA successfully separated FDT fields from healthy and glaucoma eyes and identified familiar glaucomatous patterns of loss.

Original languageEnglish (US)
Title of host publicationMedical Imaging 2014
Subtitle of host publicationImage Processing
PublisherSPIE
Volume9034
ISBN (Print)9780819498274
DOIs
StatePublished - Jan 1 2014
Externally publishedYes
EventMedical Imaging 2014: Image Processing - San Diego, CA, United States
Duration: Feb 16 2014Feb 18 2014

Other

OtherMedical Imaging 2014: Image Processing
CountryUnited States
CitySan Diego, CA
Period2/16/142/18/14

Fingerprint

visual fields
machine learning
Visual Fields
Learning systems
Technology
glaucoma
Air cushion vehicles
Glaucoma
principal components analysis
Principal Component Analysis
sensitivity
Principal component analysis
Unsupervised Machine Learning
Optic Nerve Diseases
abnormalities
Optics
optics

All Science Journal Classification (ASJC) codes

  • Electronic, Optical and Magnetic Materials
  • Atomic and Molecular Physics, and Optics
  • Biomaterials
  • Radiology Nuclear Medicine and imaging

Cite this

Yousefi, S., Goldbaum, M. H., Zangwill, L. M., Medeiros, F. A., & Bowd, C. (2014). Recognizing patterns of visual field loss using unsupervised machine learning. In Medical Imaging 2014: Image Processing (Vol. 9034). [90342M] SPIE. https://doi.org/10.1117/12.2043145

Recognizing patterns of visual field loss using unsupervised machine learning. / Yousefi, Siamak; Goldbaum, Michael H.; Zangwill, Linda M.; Medeiros, Felipe A.; Bowd, Christopher.

Medical Imaging 2014: Image Processing. Vol. 9034 SPIE, 2014. 90342M.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Yousefi, S, Goldbaum, MH, Zangwill, LM, Medeiros, FA & Bowd, C 2014, Recognizing patterns of visual field loss using unsupervised machine learning. in Medical Imaging 2014: Image Processing. vol. 9034, 90342M, SPIE, Medical Imaging 2014: Image Processing, San Diego, CA, United States, 2/16/14. https://doi.org/10.1117/12.2043145
Yousefi S, Goldbaum MH, Zangwill LM, Medeiros FA, Bowd C. Recognizing patterns of visual field loss using unsupervised machine learning. In Medical Imaging 2014: Image Processing. Vol. 9034. SPIE. 2014. 90342M https://doi.org/10.1117/12.2043145
Yousefi, Siamak ; Goldbaum, Michael H. ; Zangwill, Linda M. ; Medeiros, Felipe A. ; Bowd, Christopher. / Recognizing patterns of visual field loss using unsupervised machine learning. Medical Imaging 2014: Image Processing. Vol. 9034 SPIE, 2014.
@inproceedings{8b45423b752249a9964a16e7a5f47535,
title = "Recognizing patterns of visual field loss using unsupervised machine learning",
abstract = "Glaucoma is a potentially blinding optic neuropathy that results in a decrease in visual sensitivity. Visual field abnormalities (decreased visual sensitivity on psychophysical tests) are the primary means of glaucoma diagnosis. One form of visual field testing is Frequency Doubling Technology (FDT) that tests sensitivity at 52 points within the visual field. Like other psychophysical tests used in clinical practice, FDT results yield specific patterns of defect indicative of the disease. We used Gaussian Mixture Model with Expectation Maximization (GEM), (EM is used to estimate the model parameters) to automatically separate FDT data into clusters of normal and abnormal eyes. Principal component analysis (PCA) was used to decompose each cluster into different axes (patterns). FDT measurements were obtained from 1,190 eyes with normal FDT results and 786 eyes with abnormal (i.e., glaucomatous) FDT results, recruited from a university-based, longitudinal, multi-center, clinical study on glaucoma. The GEM input was the 52-point FDT threshold sensitivities for all eyes. The optimal GEM model separated the FDT fields into 3 clusters. Cluster 1 contained 94{\%} normal fields (94{\%} specificity) and clusters 2 and 3 combined, contained 77{\%} abnormal fields (77{\%} sensitivity). For clusters 1, 2 and 3 the optimal number of PCA-identified axes were 2, 2 and 5, respectively. GEM with PCA successfully separated FDT fields from healthy and glaucoma eyes and identified familiar glaucomatous patterns of loss.",
author = "Siamak Yousefi and Goldbaum, {Michael H.} and Zangwill, {Linda M.} and Medeiros, {Felipe A.} and Christopher Bowd",
year = "2014",
month = "1",
day = "1",
doi = "10.1117/12.2043145",
language = "English (US)",
isbn = "9780819498274",
volume = "9034",
booktitle = "Medical Imaging 2014",
publisher = "SPIE",
address = "United States",

}

TY - GEN

T1 - Recognizing patterns of visual field loss using unsupervised machine learning

AU - Yousefi, Siamak

AU - Goldbaum, Michael H.

AU - Zangwill, Linda M.

AU - Medeiros, Felipe A.

AU - Bowd, Christopher

PY - 2014/1/1

Y1 - 2014/1/1

N2 - Glaucoma is a potentially blinding optic neuropathy that results in a decrease in visual sensitivity. Visual field abnormalities (decreased visual sensitivity on psychophysical tests) are the primary means of glaucoma diagnosis. One form of visual field testing is Frequency Doubling Technology (FDT) that tests sensitivity at 52 points within the visual field. Like other psychophysical tests used in clinical practice, FDT results yield specific patterns of defect indicative of the disease. We used Gaussian Mixture Model with Expectation Maximization (GEM), (EM is used to estimate the model parameters) to automatically separate FDT data into clusters of normal and abnormal eyes. Principal component analysis (PCA) was used to decompose each cluster into different axes (patterns). FDT measurements were obtained from 1,190 eyes with normal FDT results and 786 eyes with abnormal (i.e., glaucomatous) FDT results, recruited from a university-based, longitudinal, multi-center, clinical study on glaucoma. The GEM input was the 52-point FDT threshold sensitivities for all eyes. The optimal GEM model separated the FDT fields into 3 clusters. Cluster 1 contained 94% normal fields (94% specificity) and clusters 2 and 3 combined, contained 77% abnormal fields (77% sensitivity). For clusters 1, 2 and 3 the optimal number of PCA-identified axes were 2, 2 and 5, respectively. GEM with PCA successfully separated FDT fields from healthy and glaucoma eyes and identified familiar glaucomatous patterns of loss.

AB - Glaucoma is a potentially blinding optic neuropathy that results in a decrease in visual sensitivity. Visual field abnormalities (decreased visual sensitivity on psychophysical tests) are the primary means of glaucoma diagnosis. One form of visual field testing is Frequency Doubling Technology (FDT) that tests sensitivity at 52 points within the visual field. Like other psychophysical tests used in clinical practice, FDT results yield specific patterns of defect indicative of the disease. We used Gaussian Mixture Model with Expectation Maximization (GEM), (EM is used to estimate the model parameters) to automatically separate FDT data into clusters of normal and abnormal eyes. Principal component analysis (PCA) was used to decompose each cluster into different axes (patterns). FDT measurements were obtained from 1,190 eyes with normal FDT results and 786 eyes with abnormal (i.e., glaucomatous) FDT results, recruited from a university-based, longitudinal, multi-center, clinical study on glaucoma. The GEM input was the 52-point FDT threshold sensitivities for all eyes. The optimal GEM model separated the FDT fields into 3 clusters. Cluster 1 contained 94% normal fields (94% specificity) and clusters 2 and 3 combined, contained 77% abnormal fields (77% sensitivity). For clusters 1, 2 and 3 the optimal number of PCA-identified axes were 2, 2 and 5, respectively. GEM with PCA successfully separated FDT fields from healthy and glaucoma eyes and identified familiar glaucomatous patterns of loss.

UR - http://www.scopus.com/inward/record.url?scp=84902106498&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84902106498&partnerID=8YFLogxK

U2 - 10.1117/12.2043145

DO - 10.1117/12.2043145

M3 - Conference contribution

SN - 9780819498274

VL - 9034

BT - Medical Imaging 2014

PB - SPIE

ER -