TY - JOUR
T1 - Detection of Diabetic Retinopathy from Ultra-Widefield Scanning Laser Ophthalmoscope Images
T2 - A Multicenter Deep Learning Analysis
AU - Tang, Fangyao
AU - Luenam, Phoomraphee
AU - Ran, An Ran
AU - Quadeer, Ahmed Abdul
AU - Raman, Rajiv
AU - Sen, Piyali
AU - Khan, Rehana
AU - Giridhar, Anantharaman
AU - Haridas, Swathy
AU - Iglicki, Matias
AU - Zur, Dinah
AU - Loewenstein, Anat
AU - Negri, Hermino P.
AU - Szeto, Simon
AU - Lam, Bryce Ka Yau
AU - Tham, Clement C.
AU - Sivaprasad, Sobha
AU - Mckay, Matthew
AU - Cheung, Carol Y.
N1 - Publisher Copyright:
© 2021 American Academy of Ophthalmology
PY - 2021/11
Y1 - 2021/11
N2 - Purpose: To develop a deep learning (DL) system that can detect referable diabetic retinopathy (RDR) and vision-threatening diabetic retinopathy (VTDR) from images obtained on ultra-widefield scanning laser ophthalmoscope (UWF-SLO). Design: Observational, cross-sectional study. Participants: A total of 9392 UWF-SLO images of 1903 eyes from 1022 subjects with diabetes from Hong Kong, the United Kingdom, India, and Argentina. Methods: All images were labeled according to the presence or absence of RDR and the presence or absence of VTDR. Labeling was performed by retina specialists from fundus examination, according to the International Clinical Diabetic Retinopathy Disease Severity Scale. Three convolutional neural networks (ResNet50) were trained with a transfer-learning procedure for assessing gradability and identifying VTDR and RDR. External validation was performed on 4 datasets spanning different geographical regions. Main Outcome Measures: Area under the receiver operating characteristic curve (AUROC); area under the precision-recall curve (AUPRC); sensitivity, specificity, and accuracy of the DL system in gradability assessment; and detection of RDR and VTDR. Results: For gradability assessment, the system achieved an AUROC of 0.923 (95% confidence interval [CI], 0.892–0.947), sensitivity of 86.5% (95% CI, 77.6–92.8), and specificity of 82.1% (95% CI, 77.3–86.2) for the primary validation dataset, and >0.82 AUROCs, >79.6% sensitivity, and >70.4% specificity for the geographical external validation datasets. For detecting RDR and VTDR, the AUROCs were 0.981 (95% CI, 0.977–0.984) and 0.966 (95% CI, 0.961–0.971), with sensitivities of 94.9% (95% CI, 92.3–97.9) and 87.2% (95% CI, 81.5–91.6), specificities of 95.1% (95% CI, 90.6–97.9) and 95.8% (95% CI, 93.3–97.6), and positive predictive values (PPVs) of 98.0% (95% CI, 96.1–99.0) and 91.1% (95% CI, 86.3–94.3) for the primary validation dataset, respectively. The AUROCs and accuracies for detecting both RDR and VTDR were >0.9% and >80%, respectively, for the geographical external validation datasets. The AUPRCs were >0.9, and sensitivities, specificities, and PPVs were >80% for the geographical external validation datasets for RDR and VTDR detection. Conclusions: The excellent performance achieved with this DL system for image quality assessment and detection of RDR and VTDR in UWF-SLO images highlights its potential as an efficient and effective diabetic retinopathy screening tool.
AB - Purpose: To develop a deep learning (DL) system that can detect referable diabetic retinopathy (RDR) and vision-threatening diabetic retinopathy (VTDR) from images obtained on ultra-widefield scanning laser ophthalmoscope (UWF-SLO). Design: Observational, cross-sectional study. Participants: A total of 9392 UWF-SLO images of 1903 eyes from 1022 subjects with diabetes from Hong Kong, the United Kingdom, India, and Argentina. Methods: All images were labeled according to the presence or absence of RDR and the presence or absence of VTDR. Labeling was performed by retina specialists from fundus examination, according to the International Clinical Diabetic Retinopathy Disease Severity Scale. Three convolutional neural networks (ResNet50) were trained with a transfer-learning procedure for assessing gradability and identifying VTDR and RDR. External validation was performed on 4 datasets spanning different geographical regions. Main Outcome Measures: Area under the receiver operating characteristic curve (AUROC); area under the precision-recall curve (AUPRC); sensitivity, specificity, and accuracy of the DL system in gradability assessment; and detection of RDR and VTDR. Results: For gradability assessment, the system achieved an AUROC of 0.923 (95% confidence interval [CI], 0.892–0.947), sensitivity of 86.5% (95% CI, 77.6–92.8), and specificity of 82.1% (95% CI, 77.3–86.2) for the primary validation dataset, and >0.82 AUROCs, >79.6% sensitivity, and >70.4% specificity for the geographical external validation datasets. For detecting RDR and VTDR, the AUROCs were 0.981 (95% CI, 0.977–0.984) and 0.966 (95% CI, 0.961–0.971), with sensitivities of 94.9% (95% CI, 92.3–97.9) and 87.2% (95% CI, 81.5–91.6), specificities of 95.1% (95% CI, 90.6–97.9) and 95.8% (95% CI, 93.3–97.6), and positive predictive values (PPVs) of 98.0% (95% CI, 96.1–99.0) and 91.1% (95% CI, 86.3–94.3) for the primary validation dataset, respectively. The AUROCs and accuracies for detecting both RDR and VTDR were >0.9% and >80%, respectively, for the geographical external validation datasets. The AUPRCs were >0.9, and sensitivities, specificities, and PPVs were >80% for the geographical external validation datasets for RDR and VTDR detection. Conclusions: The excellent performance achieved with this DL system for image quality assessment and detection of RDR and VTDR in UWF-SLO images highlights its potential as an efficient and effective diabetic retinopathy screening tool.
KW - Artificial intelligence
KW - Deep learning
KW - Diabetic retinopathy
KW - Imaging
KW - Ultra-widefield scanning laser ophthalmoscopy
UR - http://www.scopus.com/inward/record.url?scp=85103547618&partnerID=8YFLogxK
U2 - 10.1016/j.oret.2021.01.013
DO - 10.1016/j.oret.2021.01.013
M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???
C2 - 33540169
AN - SCOPUS:85103547618
VL - 5
SP - 1097
EP - 1106
JO - Ophthalmology Retina
JF - Ophthalmology Retina
SN - 2468-6530
IS - 11
ER -