A feasibility study for objective evaluation of visual acuity based on pattern-reversal visual evoked potentials and other related visual parameters with machine learning algorithm.
Aircrew
Machine learning
Refractive error
Visual acuity
Visual-evoked potentials
Journal
BMC ophthalmology
ISSN: 1471-2415
Titre abrégé: BMC Ophthalmol
Pays: England
ID NLM: 100967802
Informations de publication
Date de publication:
27 Jun 2023
27 Jun 2023
Historique:
received:
13
09
2022
accepted:
14
06
2023
medline:
29
6
2023
pubmed:
28
6
2023
entrez:
27
6
2023
Statut:
epublish
Résumé
To develop machine learning models for objectively evaluating visual acuity (VA) based on pattern-reversal visual evoked potentials (PRVEPs) and other related visual parameters. Twenty-four volunteers were recruited and forty-eight eyes were divided into four groups of 1.0, 0.8, 0.6, and 0.4 (decimal vision). The relationship between VA, peak time, or amplitude of P100 recorded at 5.7°, 2.6°, 1°, 34', 15', and 7' check sizes were analyzed using repeated-measures analysis of variance. Correlations between VA and P100, contrast sensitivity (CS), refractive error, wavefront aberrations, and visual field were analyzed by rank correlation. Based on meaningful P100 peak time, P100 amplitude, and other related visual parameters, four machine learning algorithms and an ensemble classification algorithm were used to construct objective assessment models for VA. Receiver operating characteristic (ROC) curves were used to compare the efficacy of different models by repeated sampling comparisons and ten-fold cross-validation. The main effects of P100 peak time and amplitude between different VA and check sizes were statistically significant (all P < 0.05). Except amplitude at 2.6° and 5.7°, VA was negatively correlated with peak time and positively correlated with amplitude. The peak time initially shortened with increasing check size and gradually lengthened after the minimum value was reached at 1°. At the 1° check size, there were statistically significant differences when comparing the peak times between the vision groups with each other (all P < 0.05), and the amplitudes of the vision reduction groups were significantly lower than that of the 1.0 vision group (all P < 0.01). The correlations between peak time, amplitude, and visual acuity were all highest at 1° (r Machine learning models established by meaning visual parameters related to visual acuity can assist in the objective evaluation of VA.
Sections du résumé
BACKGROUND
BACKGROUND
To develop machine learning models for objectively evaluating visual acuity (VA) based on pattern-reversal visual evoked potentials (PRVEPs) and other related visual parameters.
METHODS
METHODS
Twenty-four volunteers were recruited and forty-eight eyes were divided into four groups of 1.0, 0.8, 0.6, and 0.4 (decimal vision). The relationship between VA, peak time, or amplitude of P100 recorded at 5.7°, 2.6°, 1°, 34', 15', and 7' check sizes were analyzed using repeated-measures analysis of variance. Correlations between VA and P100, contrast sensitivity (CS), refractive error, wavefront aberrations, and visual field were analyzed by rank correlation. Based on meaningful P100 peak time, P100 amplitude, and other related visual parameters, four machine learning algorithms and an ensemble classification algorithm were used to construct objective assessment models for VA. Receiver operating characteristic (ROC) curves were used to compare the efficacy of different models by repeated sampling comparisons and ten-fold cross-validation.
RESULTS
RESULTS
The main effects of P100 peak time and amplitude between different VA and check sizes were statistically significant (all P < 0.05). Except amplitude at 2.6° and 5.7°, VA was negatively correlated with peak time and positively correlated with amplitude. The peak time initially shortened with increasing check size and gradually lengthened after the minimum value was reached at 1°. At the 1° check size, there were statistically significant differences when comparing the peak times between the vision groups with each other (all P < 0.05), and the amplitudes of the vision reduction groups were significantly lower than that of the 1.0 vision group (all P < 0.01). The correlations between peak time, amplitude, and visual acuity were all highest at 1° (r
CONCLUSIONS
CONCLUSIONS
Machine learning models established by meaning visual parameters related to visual acuity can assist in the objective evaluation of VA.
Identifiants
pubmed: 37369996
doi: 10.1186/s12886-023-03044-7
pii: 10.1186/s12886-023-03044-7
pmc: PMC10303860
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
293Subventions
Organisme : Medicine Promotion Project of Air Force Medical University
ID : 2018HKZL04, 2020ZTB01 and 2021JSTS28
Organisme : Medicine Promotion Project of Air Force Medical University
ID : 2018HKZL04, 2020ZTB01 and 2021JSTS28
Organisme : Medicine Promotion Project of Air Force Medical University
ID : 2018HKZL04, 2020ZTB01 and 2021JSTS28
Organisme : Aviation Medicine Project of Xijing Hospital
ID : 2023
Informations de copyright
© 2023. The Author(s).
Références
Electroencephalogr Clin Neurophysiol. 1991 May-Jun;80(3):161-6
pubmed: 1713147
Ophthalmology. 2018 Jul;125(7):1028-1036
pubmed: 29454659
Doc Ophthalmol. 2021 Feb;142(1):25-74
pubmed: 32488810
Front Cell Dev Biol. 2021 May 26;9:652848
pubmed: 34124042
Am J Ophthalmol. 1989 Nov 15;108(5):572-7
pubmed: 2817056
Doc Ophthalmol. 2017 Aug;135(1):53-67
pubmed: 28560498
Curr Opin Ophthalmol. 2006 Feb;17(1):19-26
pubmed: 16436920
J Vis. 2016 Dec 1;16(15):15
pubmed: 28006065
Front Med (Lausanne). 2021 Dec 03;8:771713
pubmed: 34926514
BMC Neurol. 2020 Mar 21;20(1):105
pubmed: 32199461
Neurology. 2021 Jan 26;96(4):e482-e490
pubmed: 33328323
J Neuroophthalmol. 2018 Sep;38(3):350-357
pubmed: 30106802
Mult Scler. 2013 Feb;19(2):173-8
pubmed: 22641300
Mult Scler Relat Disord. 2020 Jun;41:101988
pubmed: 32092503
Ophthalmology. 2001 Jan;108(1):76-80; discussion 80-1
pubmed: 11150267
J Glaucoma. 2003 Apr;12(2):134-8
pubmed: 12671468
Curr Med Sci. 2018 Apr;38(2):342-348
pubmed: 30074195
Mult Scler. 2014 Apr;20(5):617-20
pubmed: 24009163
Optom Vis Sci. 2010 May;87(5):300-12
pubmed: 20351600
J Refract Surg. 2010 Jan;26(1):52-6
pubmed: 20199013
Sensors (Basel). 2020 Sep 28;20(19):
pubmed: 32998208
BMC Ophthalmol. 2012 Aug 06;12:36
pubmed: 22866948
Doc Ophthalmol. 2015 Jun;130(3):221-9
pubmed: 25618450
Ann Transl Med. 2021 May;9(10):830
pubmed: 34164464
Neurosci Bull. 2012 Dec;28(6):737-45
pubmed: 23225314
J Refract Surg. 2004 Sep-Oct;20(5):S523-7
pubmed: 15523969
Doc Ophthalmol. 2016 Aug;133(1):1-9
pubmed: 27443562
Ophthalmology. 2007 Dec;114(12):2332-7
pubmed: 17618689
Mult Scler Relat Disord. 2020 May;40:101934
pubmed: 31986426
Doc Ophthalmol. 2010 Oct;121(2):147-54
pubmed: 20676915
J Cataract Refract Surg. 2007 Feb;33(2):210-6
pubmed: 17276260
Doc Ophthalmol. 2019 Oct;139(2):113-122
pubmed: 31187346