Comparison of Machine Learning Approaches to Improve Diagnosis of Optic Neuropathy Using Photopic Negative Response Measured Using a Handheld Device

The photopic negative response of the full-field electroretinogram (ERG) is reduced in optic neuropathies. However, technical requirements for measurement and poor classification performance have limited widespread clinical application. Recent advances in hardware facilitate efficient clinic-based r...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Tina Diao, Fareshta Kushzad, Megh D. Patel, Megha P. Bindiganavale, Munam Wasi, Mykel J. Kochenderfer, Heather E. Moss
Formato: article
Lenguaje:EN
Publicado: Frontiers Media S.A. 2021
Materias:
Acceso en línea:https://doaj.org/article/9b6f57861e6740ffb68fc7c9b5595ad8
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:9b6f57861e6740ffb68fc7c9b5595ad8
record_format dspace
spelling oai:doaj.org-article:9b6f57861e6740ffb68fc7c9b5595ad82021-12-03T05:12:14ZComparison of Machine Learning Approaches to Improve Diagnosis of Optic Neuropathy Using Photopic Negative Response Measured Using a Handheld Device2296-858X10.3389/fmed.2021.771713https://doaj.org/article/9b6f57861e6740ffb68fc7c9b5595ad82021-12-01T00:00:00Zhttps://www.frontiersin.org/articles/10.3389/fmed.2021.771713/fullhttps://doaj.org/toc/2296-858XThe photopic negative response of the full-field electroretinogram (ERG) is reduced in optic neuropathies. However, technical requirements for measurement and poor classification performance have limited widespread clinical application. Recent advances in hardware facilitate efficient clinic-based recording of the full-field ERG. Time series classification, a machine learning approach, may improve classification by using the entire ERG waveform as the input. In this study, full-field ERGs were recorded in 217 eyes (109 optic neuropathy and 108 controls) of 155 subjects. User-defined ERG features including photopic negative response were reduced in optic neuropathy eyes (p < 0.0005, generalized estimating equation models accounting for age). However, classification of optic neuropathy based on user-defined features was only fair with receiver operating characteristic area under the curve ranging between 0.62 and 0.68 and F1 score at the optimal cutoff ranging between 0.30 and 0.33. In comparison, machine learning classifiers using a variety of time series analysis approaches had F1 scores of 0.58–0.76 on a test data set. Time series classifications are promising for improving optic neuropathy diagnosis using ERG waveforms. Larger sample sizes will be important to refine the models.Tina DiaoFareshta KushzadMegh D. PatelMegha P. BindiganavaleMunam WasiMykel J. KochenderferMykel J. KochenderferHeather E. MossHeather E. MossFrontiers Media S.A.articlephotopic negative response (PhNR)electroretinogram (ERG)optic neuropathyclassificationmachine learningtime series analysisMedicine (General)R5-920ENFrontiers in Medicine, Vol 8 (2021)
institution DOAJ
collection DOAJ
language EN
topic photopic negative response (PhNR)
electroretinogram (ERG)
optic neuropathy
classification
machine learning
time series analysis
Medicine (General)
R5-920
spellingShingle photopic negative response (PhNR)
electroretinogram (ERG)
optic neuropathy
classification
machine learning
time series analysis
Medicine (General)
R5-920
Tina Diao
Fareshta Kushzad
Megh D. Patel
Megha P. Bindiganavale
Munam Wasi
Mykel J. Kochenderfer
Mykel J. Kochenderfer
Heather E. Moss
Heather E. Moss
Comparison of Machine Learning Approaches to Improve Diagnosis of Optic Neuropathy Using Photopic Negative Response Measured Using a Handheld Device
description The photopic negative response of the full-field electroretinogram (ERG) is reduced in optic neuropathies. However, technical requirements for measurement and poor classification performance have limited widespread clinical application. Recent advances in hardware facilitate efficient clinic-based recording of the full-field ERG. Time series classification, a machine learning approach, may improve classification by using the entire ERG waveform as the input. In this study, full-field ERGs were recorded in 217 eyes (109 optic neuropathy and 108 controls) of 155 subjects. User-defined ERG features including photopic negative response were reduced in optic neuropathy eyes (p < 0.0005, generalized estimating equation models accounting for age). However, classification of optic neuropathy based on user-defined features was only fair with receiver operating characteristic area under the curve ranging between 0.62 and 0.68 and F1 score at the optimal cutoff ranging between 0.30 and 0.33. In comparison, machine learning classifiers using a variety of time series analysis approaches had F1 scores of 0.58–0.76 on a test data set. Time series classifications are promising for improving optic neuropathy diagnosis using ERG waveforms. Larger sample sizes will be important to refine the models.
format article
author Tina Diao
Fareshta Kushzad
Megh D. Patel
Megha P. Bindiganavale
Munam Wasi
Mykel J. Kochenderfer
Mykel J. Kochenderfer
Heather E. Moss
Heather E. Moss
author_facet Tina Diao
Fareshta Kushzad
Megh D. Patel
Megha P. Bindiganavale
Munam Wasi
Mykel J. Kochenderfer
Mykel J. Kochenderfer
Heather E. Moss
Heather E. Moss
author_sort Tina Diao
title Comparison of Machine Learning Approaches to Improve Diagnosis of Optic Neuropathy Using Photopic Negative Response Measured Using a Handheld Device
title_short Comparison of Machine Learning Approaches to Improve Diagnosis of Optic Neuropathy Using Photopic Negative Response Measured Using a Handheld Device
title_full Comparison of Machine Learning Approaches to Improve Diagnosis of Optic Neuropathy Using Photopic Negative Response Measured Using a Handheld Device
title_fullStr Comparison of Machine Learning Approaches to Improve Diagnosis of Optic Neuropathy Using Photopic Negative Response Measured Using a Handheld Device
title_full_unstemmed Comparison of Machine Learning Approaches to Improve Diagnosis of Optic Neuropathy Using Photopic Negative Response Measured Using a Handheld Device
title_sort comparison of machine learning approaches to improve diagnosis of optic neuropathy using photopic negative response measured using a handheld device
publisher Frontiers Media S.A.
publishDate 2021
url https://doaj.org/article/9b6f57861e6740ffb68fc7c9b5595ad8
work_keys_str_mv AT tinadiao comparisonofmachinelearningapproachestoimprovediagnosisofopticneuropathyusingphotopicnegativeresponsemeasuredusingahandhelddevice
AT fareshtakushzad comparisonofmachinelearningapproachestoimprovediagnosisofopticneuropathyusingphotopicnegativeresponsemeasuredusingahandhelddevice
AT meghdpatel comparisonofmachinelearningapproachestoimprovediagnosisofopticneuropathyusingphotopicnegativeresponsemeasuredusingahandhelddevice
AT meghapbindiganavale comparisonofmachinelearningapproachestoimprovediagnosisofopticneuropathyusingphotopicnegativeresponsemeasuredusingahandhelddevice
AT munamwasi comparisonofmachinelearningapproachestoimprovediagnosisofopticneuropathyusingphotopicnegativeresponsemeasuredusingahandhelddevice
AT mykeljkochenderfer comparisonofmachinelearningapproachestoimprovediagnosisofopticneuropathyusingphotopicnegativeresponsemeasuredusingahandhelddevice
AT mykeljkochenderfer comparisonofmachinelearningapproachestoimprovediagnosisofopticneuropathyusingphotopicnegativeresponsemeasuredusingahandhelddevice
AT heatheremoss comparisonofmachinelearningapproachestoimprovediagnosisofopticneuropathyusingphotopicnegativeresponsemeasuredusingahandhelddevice
AT heatheremoss comparisonofmachinelearningapproachestoimprovediagnosisofopticneuropathyusingphotopicnegativeresponsemeasuredusingahandhelddevice
_version_ 1718373947581399040