Fully automated disease severity assessment and treatment monitoring in retinopathy of prematurity using deep learning

James M. Brown, John Campbell, Andrew Beers, Ken Chang, Kyra Donohue, Susan Ostmo, R. V.Paul Chan, Jennifer Dy, Deniz Erdogmus, Stratis Ioannidis, Michael Chiang, Jayashree Kalpathy-Cramer

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

2 Citations (Scopus)

Abstract

Retinopathy of prematurity (ROP) is a disease that affects premature infants, where abnormal growth of the retinal blood vessels can lead to blindness unless treated accordingly. Infants considered at risk of severe ROP are monitored for symptoms of plus disease, characterized by arterial tortuosity and venous dilation at the posterior pole, with a standard photographic definition. Disagreement among ROP experts in diagnosing plus disease has driven the development of computer-based methods that classify images based on hand-crafted features extracted from the vasculature. However, most of these approaches are semi-Automated, which are time-consuming and subject to variability. In contrast, deep learning is a fully automated approach that has shown great promise in a wide variety of domains, including medical genetics, informatics and imaging. Convolutional neural networks (CNNs) are deep networks which learn rich representations of disease features that are highly robust to variations in acquisition and image quality. In this study, we utilized a U-Net architecture to perform vessel segmentation and then a GoogLeNet to perform disease classification. The classifier was trained on 3,000 retinal images and validated on an independent test set of patients with different observed progressions and treatments. We show that our fully automated algorithm can be used to monitor the progression of plus disease over multiple patient visits with results that are consistent with the experts' consensus diagnosis. Future work will aim to further validate the method on larger cohorts of patients to assess its applicability within the clinic as a treatment monitoring tool.

Original languageEnglish (US)
Title of host publicationMedical Imaging 2018
Subtitle of host publicationImaging Informatics for Healthcare, Research, and Applications
PublisherSPIE
Volume10579
ISBN (Electronic)9781510616479
DOIs
StatePublished - Jan 1 2018
EventMedical Imaging 2018: Imaging Informatics for Healthcare, Research, and Applications - Houston, United States
Duration: Feb 13 2018Feb 15 2018

Other

OtherMedical Imaging 2018: Imaging Informatics for Healthcare, Research, and Applications
CountryUnited States
CityHouston
Period2/13/182/15/18

Fingerprint

Retinopathy of Prematurity
learning
Learning
Monitoring
Infant, Premature, Diseases
progressions
Medical Informatics
Retinal Vessels
Medical Genetics
Diagnostic Imaging
Blindness
Therapeutics
blindness
retinal images
Disease Progression
Dilatation
Hand
blood vessels
Blood vessels
classifiers

Keywords

  • Deep learning
  • Machine learning
  • Plus disease
  • Retina
  • Retinopathy of prematurity

ASJC Scopus subject areas

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

Cite this

Brown, J. M., Campbell, J., Beers, A., Chang, K., Donohue, K., Ostmo, S., ... Kalpathy-Cramer, J. (2018). Fully automated disease severity assessment and treatment monitoring in retinopathy of prematurity using deep learning. In Medical Imaging 2018: Imaging Informatics for Healthcare, Research, and Applications (Vol. 10579). [105790Q] SPIE. https://doi.org/10.1117/12.2295942

Fully automated disease severity assessment and treatment monitoring in retinopathy of prematurity using deep learning. / Brown, James M.; Campbell, John; Beers, Andrew; Chang, Ken; Donohue, Kyra; Ostmo, Susan; Chan, R. V.Paul; Dy, Jennifer; Erdogmus, Deniz; Ioannidis, Stratis; Chiang, Michael; Kalpathy-Cramer, Jayashree.

Medical Imaging 2018: Imaging Informatics for Healthcare, Research, and Applications. Vol. 10579 SPIE, 2018. 105790Q.

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

Brown, JM, Campbell, J, Beers, A, Chang, K, Donohue, K, Ostmo, S, Chan, RVP, Dy, J, Erdogmus, D, Ioannidis, S, Chiang, M & Kalpathy-Cramer, J 2018, Fully automated disease severity assessment and treatment monitoring in retinopathy of prematurity using deep learning. in Medical Imaging 2018: Imaging Informatics for Healthcare, Research, and Applications. vol. 10579, 105790Q, SPIE, Medical Imaging 2018: Imaging Informatics for Healthcare, Research, and Applications, Houston, United States, 2/13/18. https://doi.org/10.1117/12.2295942
Brown JM, Campbell J, Beers A, Chang K, Donohue K, Ostmo S et al. Fully automated disease severity assessment and treatment monitoring in retinopathy of prematurity using deep learning. In Medical Imaging 2018: Imaging Informatics for Healthcare, Research, and Applications. Vol. 10579. SPIE. 2018. 105790Q https://doi.org/10.1117/12.2295942
Brown, James M. ; Campbell, John ; Beers, Andrew ; Chang, Ken ; Donohue, Kyra ; Ostmo, Susan ; Chan, R. V.Paul ; Dy, Jennifer ; Erdogmus, Deniz ; Ioannidis, Stratis ; Chiang, Michael ; Kalpathy-Cramer, Jayashree. / Fully automated disease severity assessment and treatment monitoring in retinopathy of prematurity using deep learning. Medical Imaging 2018: Imaging Informatics for Healthcare, Research, and Applications. Vol. 10579 SPIE, 2018.
@inproceedings{5f68786df7e04d90a4c7473e5159046c,
title = "Fully automated disease severity assessment and treatment monitoring in retinopathy of prematurity using deep learning",
abstract = "Retinopathy of prematurity (ROP) is a disease that affects premature infants, where abnormal growth of the retinal blood vessels can lead to blindness unless treated accordingly. Infants considered at risk of severe ROP are monitored for symptoms of plus disease, characterized by arterial tortuosity and venous dilation at the posterior pole, with a standard photographic definition. Disagreement among ROP experts in diagnosing plus disease has driven the development of computer-based methods that classify images based on hand-crafted features extracted from the vasculature. However, most of these approaches are semi-Automated, which are time-consuming and subject to variability. In contrast, deep learning is a fully automated approach that has shown great promise in a wide variety of domains, including medical genetics, informatics and imaging. Convolutional neural networks (CNNs) are deep networks which learn rich representations of disease features that are highly robust to variations in acquisition and image quality. In this study, we utilized a U-Net architecture to perform vessel segmentation and then a GoogLeNet to perform disease classification. The classifier was trained on 3,000 retinal images and validated on an independent test set of patients with different observed progressions and treatments. We show that our fully automated algorithm can be used to monitor the progression of plus disease over multiple patient visits with results that are consistent with the experts' consensus diagnosis. Future work will aim to further validate the method on larger cohorts of patients to assess its applicability within the clinic as a treatment monitoring tool.",
keywords = "Deep learning, Machine learning, Plus disease, Retina, Retinopathy of prematurity",
author = "Brown, {James M.} and John Campbell and Andrew Beers and Ken Chang and Kyra Donohue and Susan Ostmo and Chan, {R. V.Paul} and Jennifer Dy and Deniz Erdogmus and Stratis Ioannidis and Michael Chiang and Jayashree Kalpathy-Cramer",
year = "2018",
month = "1",
day = "1",
doi = "10.1117/12.2295942",
language = "English (US)",
volume = "10579",
booktitle = "Medical Imaging 2018",
publisher = "SPIE",

}

TY - GEN

T1 - Fully automated disease severity assessment and treatment monitoring in retinopathy of prematurity using deep learning

AU - Brown, James M.

AU - Campbell, John

AU - Beers, Andrew

AU - Chang, Ken

AU - Donohue, Kyra

AU - Ostmo, Susan

AU - Chan, R. V.Paul

AU - Dy, Jennifer

AU - Erdogmus, Deniz

AU - Ioannidis, Stratis

AU - Chiang, Michael

AU - Kalpathy-Cramer, Jayashree

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Retinopathy of prematurity (ROP) is a disease that affects premature infants, where abnormal growth of the retinal blood vessels can lead to blindness unless treated accordingly. Infants considered at risk of severe ROP are monitored for symptoms of plus disease, characterized by arterial tortuosity and venous dilation at the posterior pole, with a standard photographic definition. Disagreement among ROP experts in diagnosing plus disease has driven the development of computer-based methods that classify images based on hand-crafted features extracted from the vasculature. However, most of these approaches are semi-Automated, which are time-consuming and subject to variability. In contrast, deep learning is a fully automated approach that has shown great promise in a wide variety of domains, including medical genetics, informatics and imaging. Convolutional neural networks (CNNs) are deep networks which learn rich representations of disease features that are highly robust to variations in acquisition and image quality. In this study, we utilized a U-Net architecture to perform vessel segmentation and then a GoogLeNet to perform disease classification. The classifier was trained on 3,000 retinal images and validated on an independent test set of patients with different observed progressions and treatments. We show that our fully automated algorithm can be used to monitor the progression of plus disease over multiple patient visits with results that are consistent with the experts' consensus diagnosis. Future work will aim to further validate the method on larger cohorts of patients to assess its applicability within the clinic as a treatment monitoring tool.

AB - Retinopathy of prematurity (ROP) is a disease that affects premature infants, where abnormal growth of the retinal blood vessels can lead to blindness unless treated accordingly. Infants considered at risk of severe ROP are monitored for symptoms of plus disease, characterized by arterial tortuosity and venous dilation at the posterior pole, with a standard photographic definition. Disagreement among ROP experts in diagnosing plus disease has driven the development of computer-based methods that classify images based on hand-crafted features extracted from the vasculature. However, most of these approaches are semi-Automated, which are time-consuming and subject to variability. In contrast, deep learning is a fully automated approach that has shown great promise in a wide variety of domains, including medical genetics, informatics and imaging. Convolutional neural networks (CNNs) are deep networks which learn rich representations of disease features that are highly robust to variations in acquisition and image quality. In this study, we utilized a U-Net architecture to perform vessel segmentation and then a GoogLeNet to perform disease classification. The classifier was trained on 3,000 retinal images and validated on an independent test set of patients with different observed progressions and treatments. We show that our fully automated algorithm can be used to monitor the progression of plus disease over multiple patient visits with results that are consistent with the experts' consensus diagnosis. Future work will aim to further validate the method on larger cohorts of patients to assess its applicability within the clinic as a treatment monitoring tool.

KW - Deep learning

KW - Machine learning

KW - Plus disease

KW - Retina

KW - Retinopathy of prematurity

UR - http://www.scopus.com/inward/record.url?scp=85047729156&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85047729156&partnerID=8YFLogxK

U2 - 10.1117/12.2295942

DO - 10.1117/12.2295942

M3 - Conference contribution

AN - SCOPUS:85047729156

VL - 10579

BT - Medical Imaging 2018

PB - SPIE

ER -