Clinical validation of a smartphone-based adapter for optic disc imaging in Kenya

Andrew Bastawrous, Mario Ettore Giardini, Nigel M. Bolster, Tunde Peto, Nisha Shah, Iain A.T. Livingstone, Helen A. Weiss, Sen Hu, Hillary Rono, Hannah Kuper, Matthew Burton

Research output: Contribution to journalArticle

35 Citations (Scopus)

Abstract

Visualization and interpretation of the optic nerve and retina are essential parts of most physical examinations. To design and validate a smartphone-based retinal adapter enabling image capture and remote grading of the retina. This validation study compared the grading of optic nerves from smartphone images with those of a digital retinal camera. Both image sets were independently graded at Moorfields Eye Hospital Reading Centre. Nested within the 6-year follow-up (January 7, 2013, to March 12, 2014) of the Nakuru Eye Disease Cohort in Kenya, 1460 adults (2920 eyes) 55 years and older were recruited consecutively from the study. A subset of 100 optic disc images from both methods were further used to validate a grading app for the optic nerves. Data analysis was performed April 7 to April 12, 2015. Vertical cup-disc ratio for each testwas compared in terms of agreement (Bland-Altman and weighted κ) and test-retest variability. A total of 2152 optic nerve images were available from both methods (also 371 from the reference camera but not the smartphone, 170 from the smartphone but not the reference camera, and 227 from neither the reference camera nor the smartphone). Bland-Altman analysis revealed a mean difference of 0.02 (95%CI, −0.21 to 0.17) and a weighted κ coefficient of 0.69 (excellent agreement). The grades of an experienced retinal photographer were compared with those of a lay photographer (no health care experience before the study), and no observable difference in image acquisition quality was found. Nonclinical photographers using the low-cost smartphone adapter were able to acquire optic nerve images at a standard that enabled independent remote grading of the images comparable to those acquired using a desktop retinal camera operated by an ophthalmic assistant. The potential for task shifting and the detection of avoidable causes of blindness in the most at-risk communities makes this an attractive public health intervention.
LanguageEnglish
Pages151-158
Number of pages8
JournalJAMA Ophthalmology
Volume134
Issue number2
Early online date25 Nov 2015
DOIs
Publication statusPublished - 12 Feb 2016

Fingerprint

Kenya
Smartphones
Optic Disk
Optic Nerve
Optics
Imaging techniques
Cameras
Ophthalmic Assistants
Retina
Eye Diseases
Validation Studies
Image acquisition
Digital cameras
Public health
Blindness
Health care
Application programs
Physical Examination
Smartphone
Reading

Keywords

  • optic disc imaging
  • Kenya
  • retinal image capture
  • smartphone
  • optic nerve

Cite this

Bastawrous, Andrew ; Giardini, Mario Ettore ; Bolster, Nigel M. ; Peto, Tunde ; Shah, Nisha ; Livingstone, Iain A.T. ; Weiss, Helen A. ; Hu, Sen ; Rono, Hillary ; Kuper, Hannah ; Burton, Matthew. / Clinical validation of a smartphone-based adapter for optic disc imaging in Kenya. In: JAMA Ophthalmology. 2016 ; Vol. 134, No. 2. pp. 151-158.
@article{3d6ad9a513924a968dbdab57e8d61ffc,
title = "Clinical validation of a smartphone-based adapter for optic disc imaging in Kenya",
abstract = "Visualization and interpretation of the optic nerve and retina are essential parts of most physical examinations. To design and validate a smartphone-based retinal adapter enabling image capture and remote grading of the retina. This validation study compared the grading of optic nerves from smartphone images with those of a digital retinal camera. Both image sets were independently graded at Moorfields Eye Hospital Reading Centre. Nested within the 6-year follow-up (January 7, 2013, to March 12, 2014) of the Nakuru Eye Disease Cohort in Kenya, 1460 adults (2920 eyes) 55 years and older were recruited consecutively from the study. A subset of 100 optic disc images from both methods were further used to validate a grading app for the optic nerves. Data analysis was performed April 7 to April 12, 2015. Vertical cup-disc ratio for each testwas compared in terms of agreement (Bland-Altman and weighted κ) and test-retest variability. A total of 2152 optic nerve images were available from both methods (also 371 from the reference camera but not the smartphone, 170 from the smartphone but not the reference camera, and 227 from neither the reference camera nor the smartphone). Bland-Altman analysis revealed a mean difference of 0.02 (95{\%}CI, −0.21 to 0.17) and a weighted κ coefficient of 0.69 (excellent agreement). The grades of an experienced retinal photographer were compared with those of a lay photographer (no health care experience before the study), and no observable difference in image acquisition quality was found. Nonclinical photographers using the low-cost smartphone adapter were able to acquire optic nerve images at a standard that enabled independent remote grading of the images comparable to those acquired using a desktop retinal camera operated by an ophthalmic assistant. The potential for task shifting and the detection of avoidable causes of blindness in the most at-risk communities makes this an attractive public health intervention.",
keywords = "optic disc imaging, Kenya, retinal image capture, smartphone, optic nerve",
author = "Andrew Bastawrous and Giardini, {Mario Ettore} and Bolster, {Nigel M.} and Tunde Peto and Nisha Shah and Livingstone, {Iain A.T.} and Weiss, {Helen A.} and Sen Hu and Hillary Rono and Hannah Kuper and Matthew Burton",
year = "2016",
month = "2",
day = "12",
doi = "10.1001/jamaophthalmol.2015.4625",
language = "English",
volume = "134",
pages = "151--158",
journal = "JAMA Ophthalmology",
issn = "2168-6165",
publisher = "American Medical Association",
number = "2",

}

Bastawrous, A, Giardini, ME, Bolster, NM, Peto, T, Shah, N, Livingstone, IAT, Weiss, HA, Hu, S, Rono, H, Kuper, H & Burton, M 2016, 'Clinical validation of a smartphone-based adapter for optic disc imaging in Kenya' JAMA Ophthalmology, vol. 134, no. 2, pp. 151-158. https://doi.org/10.1001/jamaophthalmol.2015.4625

Clinical validation of a smartphone-based adapter for optic disc imaging in Kenya. / Bastawrous, Andrew; Giardini, Mario Ettore; Bolster, Nigel M.; Peto, Tunde; Shah, Nisha; Livingstone, Iain A.T.; Weiss, Helen A.; Hu, Sen; Rono, Hillary; Kuper, Hannah; Burton, Matthew.

In: JAMA Ophthalmology, Vol. 134, No. 2, 12.02.2016, p. 151-158.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Clinical validation of a smartphone-based adapter for optic disc imaging in Kenya

AU - Bastawrous, Andrew

AU - Giardini, Mario Ettore

AU - Bolster, Nigel M.

AU - Peto, Tunde

AU - Shah, Nisha

AU - Livingstone, Iain A.T.

AU - Weiss, Helen A.

AU - Hu, Sen

AU - Rono, Hillary

AU - Kuper, Hannah

AU - Burton, Matthew

PY - 2016/2/12

Y1 - 2016/2/12

N2 - Visualization and interpretation of the optic nerve and retina are essential parts of most physical examinations. To design and validate a smartphone-based retinal adapter enabling image capture and remote grading of the retina. This validation study compared the grading of optic nerves from smartphone images with those of a digital retinal camera. Both image sets were independently graded at Moorfields Eye Hospital Reading Centre. Nested within the 6-year follow-up (January 7, 2013, to March 12, 2014) of the Nakuru Eye Disease Cohort in Kenya, 1460 adults (2920 eyes) 55 years and older were recruited consecutively from the study. A subset of 100 optic disc images from both methods were further used to validate a grading app for the optic nerves. Data analysis was performed April 7 to April 12, 2015. Vertical cup-disc ratio for each testwas compared in terms of agreement (Bland-Altman and weighted κ) and test-retest variability. A total of 2152 optic nerve images were available from both methods (also 371 from the reference camera but not the smartphone, 170 from the smartphone but not the reference camera, and 227 from neither the reference camera nor the smartphone). Bland-Altman analysis revealed a mean difference of 0.02 (95%CI, −0.21 to 0.17) and a weighted κ coefficient of 0.69 (excellent agreement). The grades of an experienced retinal photographer were compared with those of a lay photographer (no health care experience before the study), and no observable difference in image acquisition quality was found. Nonclinical photographers using the low-cost smartphone adapter were able to acquire optic nerve images at a standard that enabled independent remote grading of the images comparable to those acquired using a desktop retinal camera operated by an ophthalmic assistant. The potential for task shifting and the detection of avoidable causes of blindness in the most at-risk communities makes this an attractive public health intervention.

AB - Visualization and interpretation of the optic nerve and retina are essential parts of most physical examinations. To design and validate a smartphone-based retinal adapter enabling image capture and remote grading of the retina. This validation study compared the grading of optic nerves from smartphone images with those of a digital retinal camera. Both image sets were independently graded at Moorfields Eye Hospital Reading Centre. Nested within the 6-year follow-up (January 7, 2013, to March 12, 2014) of the Nakuru Eye Disease Cohort in Kenya, 1460 adults (2920 eyes) 55 years and older were recruited consecutively from the study. A subset of 100 optic disc images from both methods were further used to validate a grading app for the optic nerves. Data analysis was performed April 7 to April 12, 2015. Vertical cup-disc ratio for each testwas compared in terms of agreement (Bland-Altman and weighted κ) and test-retest variability. A total of 2152 optic nerve images were available from both methods (also 371 from the reference camera but not the smartphone, 170 from the smartphone but not the reference camera, and 227 from neither the reference camera nor the smartphone). Bland-Altman analysis revealed a mean difference of 0.02 (95%CI, −0.21 to 0.17) and a weighted κ coefficient of 0.69 (excellent agreement). The grades of an experienced retinal photographer were compared with those of a lay photographer (no health care experience before the study), and no observable difference in image acquisition quality was found. Nonclinical photographers using the low-cost smartphone adapter were able to acquire optic nerve images at a standard that enabled independent remote grading of the images comparable to those acquired using a desktop retinal camera operated by an ophthalmic assistant. The potential for task shifting and the detection of avoidable causes of blindness in the most at-risk communities makes this an attractive public health intervention.

KW - optic disc imaging

KW - Kenya

KW - retinal image capture

KW - smartphone

KW - optic nerve

U2 - 10.1001/jamaophthalmol.2015.4625

DO - 10.1001/jamaophthalmol.2015.4625

M3 - Article

VL - 134

SP - 151

EP - 158

JO - JAMA Ophthalmology

T2 - JAMA Ophthalmology

JF - JAMA Ophthalmology

SN - 2168-6165

IS - 2

ER -