Crowdsourcing and Automated Retinal Image Analysis for Diabetic Retinopathy

Lucy I. Mudie; Xueyang Wang; David S. Friedman; Christopher J. Brady

doi:10.1007/s11892-017-0940-x

Crowdsourcing and Automated Retinal Image Analysis for Diabetic Retinopathy

Lucy I. Mudie, Xueyang Wang, David S. Friedman, Christopher J. Brady

School of Medicine

Research output: Contribution to journal › Review article › peer-review

10 Scopus citations

Abstract

Purpose of Review: As the number of people with diabetic retinopathy (DR) in the USA is expected to increase threefold by 2050, the need to reduce health care costs associated with screening for this treatable disease is ever present. Crowdsourcing and automated retinal image analysis (ARIA) are two areas where new technology has been applied to reduce costs in screening for DR. This paper reviews the current literature surrounding these new technologies. Recent Findings: Crowdsourcing has high sensitivity for normal vs abnormal images; however, when multiple categories for severity of DR are added, specificity is reduced. ARIAs have higher sensitivity and specificity, and some commercial ARIA programs are already in use. Deep learning enhanced ARIAs appear to offer even more improvement in ARIA grading accuracy. Summary: The utilization of crowdsourcing and ARIAs may be a key to reducing the time and cost burden of processing images from DR screening.

Original language	English (US)
Article number	106
Journal	Current diabetes reports
Volume	17
Issue number	11
DOIs	https://doi.org/10.1007/s11892-017-0940-x
State	Published - Nov 1 2017

Keywords

Amazon Mechanical Turk
Automated retinal image analysis
Crowdsourcing
Diabetic retinopathy
Telemedicine

ASJC Scopus subject areas

Internal Medicine
Endocrinology, Diabetes and Metabolism

Access to Document

10.1007/s11892-017-0940-x

Cite this

@article{14ae76b653764d35aef1a2215fb73aa6,

title = "Crowdsourcing and Automated Retinal Image Analysis for Diabetic Retinopathy",

abstract = "Purpose of Review: As the number of people with diabetic retinopathy (DR) in the USA is expected to increase threefold by 2050, the need to reduce health care costs associated with screening for this treatable disease is ever present. Crowdsourcing and automated retinal image analysis (ARIA) are two areas where new technology has been applied to reduce costs in screening for DR. This paper reviews the current literature surrounding these new technologies. Recent Findings: Crowdsourcing has high sensitivity for normal vs abnormal images; however, when multiple categories for severity of DR are added, specificity is reduced. ARIAs have higher sensitivity and specificity, and some commercial ARIA programs are already in use. Deep learning enhanced ARIAs appear to offer even more improvement in ARIA grading accuracy. Summary: The utilization of crowdsourcing and ARIAs may be a key to reducing the time and cost burden of processing images from DR screening.",

keywords = "Amazon Mechanical Turk, Automated retinal image analysis, Crowdsourcing, Diabetic retinopathy, Telemedicine",

author = "Mudie, {Lucy I.} and Xueyang Wang and Friedman, {David S.} and Brady, {Christopher J.}",

note = "Funding Information: Funding Information This publication was made possible by the Johns Hopkins Institute for Clinical and Translational Research (ICTR) which is funded in part by Grant Number KL2TR001077 from the National Center for Advancing Translational Sciences (NCATS), a component of the National Institutes of Health (NIH), and NIH Roadmap for Medical Research, and by the Johns Hopkins-University of Maryland Diabetes Research Center Pilot and Feasibility Award under National Institute of Diabetes and Digestion P30DK079637 (Hussain, PI). Publisher Copyright: {\textcopyright} 2017, Springer Science+Business Media, LLC.",

year = "2017",

month = nov,

day = "1",

doi = "10.1007/s11892-017-0940-x",

language = "English (US)",

volume = "17",

journal = "Current diabetes reports",

issn = "1534-4827",

publisher = "Current Medicine Group",

number = "11",

}

TY - JOUR

T1 - Crowdsourcing and Automated Retinal Image Analysis for Diabetic Retinopathy

AU - Mudie, Lucy I.

AU - Wang, Xueyang

AU - Friedman, David S.

AU - Brady, Christopher J.

N1 - Funding Information: Funding Information This publication was made possible by the Johns Hopkins Institute for Clinical and Translational Research (ICTR) which is funded in part by Grant Number KL2TR001077 from the National Center for Advancing Translational Sciences (NCATS), a component of the National Institutes of Health (NIH), and NIH Roadmap for Medical Research, and by the Johns Hopkins-University of Maryland Diabetes Research Center Pilot and Feasibility Award under National Institute of Diabetes and Digestion P30DK079637 (Hussain, PI). Publisher Copyright: © 2017, Springer Science+Business Media, LLC.

PY - 2017/11/1

Y1 - 2017/11/1

N2 - Purpose of Review: As the number of people with diabetic retinopathy (DR) in the USA is expected to increase threefold by 2050, the need to reduce health care costs associated with screening for this treatable disease is ever present. Crowdsourcing and automated retinal image analysis (ARIA) are two areas where new technology has been applied to reduce costs in screening for DR. This paper reviews the current literature surrounding these new technologies. Recent Findings: Crowdsourcing has high sensitivity for normal vs abnormal images; however, when multiple categories for severity of DR are added, specificity is reduced. ARIAs have higher sensitivity and specificity, and some commercial ARIA programs are already in use. Deep learning enhanced ARIAs appear to offer even more improvement in ARIA grading accuracy. Summary: The utilization of crowdsourcing and ARIAs may be a key to reducing the time and cost burden of processing images from DR screening.

AB - Purpose of Review: As the number of people with diabetic retinopathy (DR) in the USA is expected to increase threefold by 2050, the need to reduce health care costs associated with screening for this treatable disease is ever present. Crowdsourcing and automated retinal image analysis (ARIA) are two areas where new technology has been applied to reduce costs in screening for DR. This paper reviews the current literature surrounding these new technologies. Recent Findings: Crowdsourcing has high sensitivity for normal vs abnormal images; however, when multiple categories for severity of DR are added, specificity is reduced. ARIAs have higher sensitivity and specificity, and some commercial ARIA programs are already in use. Deep learning enhanced ARIAs appear to offer even more improvement in ARIA grading accuracy. Summary: The utilization of crowdsourcing and ARIAs may be a key to reducing the time and cost burden of processing images from DR screening.

KW - Amazon Mechanical Turk

KW - Automated retinal image analysis

KW - Crowdsourcing

KW - Diabetic retinopathy

KW - Telemedicine

UR - http://www.scopus.com/inward/record.url?scp=85029847204&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85029847204&partnerID=8YFLogxK

U2 - 10.1007/s11892-017-0940-x

DO - 10.1007/s11892-017-0940-x

M3 - Review article

C2 - 28942485

AN - SCOPUS:85029847204

SN - 1534-4827

VL - 17

JO - Current diabetes reports

JF - Current diabetes reports

IS - 11

M1 - 106

ER -

Crowdsourcing and Automated Retinal Image Analysis for Diabetic Retinopathy

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this