TY - JOUR
T1 - Evaluation of an Artificial Intelligence System for Retinopathy of Prematurity Screening in Nepal and Mongolia
AU - Cole, Emily
AU - Valikodath, Nita G.
AU - Al-Khaled, Tala
AU - Bajimaya, Sanyam
AU - KC, Sagun
AU - Chuluunbat, Tsengelmaa
AU - Munkhuu, Bayalag
AU - Jonas, Karyn E.
AU - Chuluunkhuu, Chimgee
AU - MacKeen, Leslie D.
AU - Yap, Vivien
AU - Hallak, Joelle
AU - Ostmo, Susan
AU - Wu, Wei Chi
AU - Coyner, Aaron S.
AU - Singh, Praveer
AU - Kalpathy-Cramer, Jayashree
AU - Chiang, Michael F.
AU - Campbell, J. Peter
AU - Chan, R. V.Paul
N1 - Publisher Copyright:
© 2022 American Academy of Ophthalmology
PY - 2022/12
Y1 - 2022/12
N2 - Purpose: To evaluate the performance of a deep learning (DL) algorithm for retinopathy of prematurity (ROP) screening in Nepal and Mongolia. Design: Retrospective analysis of prospectively collected clinical data. Participants: Clinical information and fundus images were obtained from infants in 2 ROP screening programs in Nepal and Mongolia. Methods: Fundus images were obtained using the Forus 3nethra neo (Forus Health) in Nepal and the RetCam Portable (Natus Medical, Inc.) in Mongolia. The overall severity of ROP was determined from the medical record using the International Classification of ROP (ICROP). The presence of plus disease was determined independently in each image using a reference standard diagnosis. The Imaging and Informatics for ROP (i-ROP) DL algorithm was trained on images from the RetCam to classify plus disease and to assign a vascular severity score (VSS) from 1 through 9. Main Outcome Measures: Area under the receiver operating characteristic curve and area under the precision-recall curve for the presence of plus disease or type 1 ROP and association between VSS and ICROP disease category. Results: The prevalence of type 1 ROP was found to be higher in Mongolia (14.0%) than in Nepal (2.2%; P < 0.001) in these data sets. In Mongolia (RetCam images), the area under the receiver operating characteristic curve for examination-level plus disease detection was 0.968, and the area under the precision-recall curve was 0.823. In Nepal (Forus images), these values were 0.999 and 0.993, respectively. The ROP VSS was associated with ICROP classification in both datasets (P < 0.001). At the population level, the median VSS was found to be higher in Mongolia (2.7; interquartile range [IQR], 1.3–5.4]) as compared with Nepal (1.9; IQR, 1.2–3.4; P < 0.001). Conclusions: These data provide preliminary evidence of the effectiveness of the i-ROP DL algorithm for ROP screening in neonatal populations in Nepal and Mongolia using multiple camera systems and are useful for consideration in future clinical implementation of artificial intelligence–based ROP screening in low- and middle-income countries.
AB - Purpose: To evaluate the performance of a deep learning (DL) algorithm for retinopathy of prematurity (ROP) screening in Nepal and Mongolia. Design: Retrospective analysis of prospectively collected clinical data. Participants: Clinical information and fundus images were obtained from infants in 2 ROP screening programs in Nepal and Mongolia. Methods: Fundus images were obtained using the Forus 3nethra neo (Forus Health) in Nepal and the RetCam Portable (Natus Medical, Inc.) in Mongolia. The overall severity of ROP was determined from the medical record using the International Classification of ROP (ICROP). The presence of plus disease was determined independently in each image using a reference standard diagnosis. The Imaging and Informatics for ROP (i-ROP) DL algorithm was trained on images from the RetCam to classify plus disease and to assign a vascular severity score (VSS) from 1 through 9. Main Outcome Measures: Area under the receiver operating characteristic curve and area under the precision-recall curve for the presence of plus disease or type 1 ROP and association between VSS and ICROP disease category. Results: The prevalence of type 1 ROP was found to be higher in Mongolia (14.0%) than in Nepal (2.2%; P < 0.001) in these data sets. In Mongolia (RetCam images), the area under the receiver operating characteristic curve for examination-level plus disease detection was 0.968, and the area under the precision-recall curve was 0.823. In Nepal (Forus images), these values were 0.999 and 0.993, respectively. The ROP VSS was associated with ICROP classification in both datasets (P < 0.001). At the population level, the median VSS was found to be higher in Mongolia (2.7; interquartile range [IQR], 1.3–5.4]) as compared with Nepal (1.9; IQR, 1.2–3.4; P < 0.001). Conclusions: These data provide preliminary evidence of the effectiveness of the i-ROP DL algorithm for ROP screening in neonatal populations in Nepal and Mongolia using multiple camera systems and are useful for consideration in future clinical implementation of artificial intelligence–based ROP screening in low- and middle-income countries.
KW - Artificial intelligence
KW - Deep learning
KW - Mongolia
KW - Nepal
KW - Retinopathy of prematurity
UR - http://www.scopus.com/inward/record.url?scp=85134414406&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85134414406&partnerID=8YFLogxK
U2 - 10.1016/j.xops.2022.100165
DO - 10.1016/j.xops.2022.100165
M3 - Article
AN - SCOPUS:85134414406
SN - 2666-9145
VL - 2
JO - Ophthalmology Science
JF - Ophthalmology Science
IS - 4
M1 - 100165
ER -