TY - CHAP
T1 - Artificial Intelligence for Pediatric Retinal Diseases
AU - Acaba-Berrocal, Luis
AU - Coyner, Aaron
AU - Chiang, Michael F.
AU - Peter Campbell, J.
AU - Paul Chan, R. V.
N1 - Publisher Copyright:
© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2023.
PY - 2023/1/1
Y1 - 2023/1/1
N2 - Artificial intelligence (AI) is an evolving field within medicine in general and specifically in ophthalmology. Ophthalmologic research and development of AI platforms for pediatric retinal disease have largely focused on retinopathy of prematurity (ROP) screening, diagnosis, and patient follow-up. Platforms such as the Imaging and Informatics in ROP Deep Learning (i-ROP DL) algorithm have demonstrated both high sensitivity and specificity for the diagnosis of plus disease. In addition, the i-ROP DL vascular severity score has been shown to be helpful not only in diagnosing ROP disease severity, but also in tracking disease progression and diagnosing treatment requiring-ROP. Therefore, AI has shown promise in helping limit screening on low-risk infants and directing resources to eyes at highest risk for treatment-requiring ROP, increasing the efficiency of ROP screenings and identifying patients likely to need treatment. AI has also shown potential in assisting real-time in surgeries. However, barriers including costs, generalizability, and medicolegal interpretations remain and need to be addressed before the algorithms become common place in clinical practice.
AB - Artificial intelligence (AI) is an evolving field within medicine in general and specifically in ophthalmology. Ophthalmologic research and development of AI platforms for pediatric retinal disease have largely focused on retinopathy of prematurity (ROP) screening, diagnosis, and patient follow-up. Platforms such as the Imaging and Informatics in ROP Deep Learning (i-ROP DL) algorithm have demonstrated both high sensitivity and specificity for the diagnosis of plus disease. In addition, the i-ROP DL vascular severity score has been shown to be helpful not only in diagnosing ROP disease severity, but also in tracking disease progression and diagnosing treatment requiring-ROP. Therefore, AI has shown promise in helping limit screening on low-risk infants and directing resources to eyes at highest risk for treatment-requiring ROP, increasing the efficiency of ROP screenings and identifying patients likely to need treatment. AI has also shown potential in assisting real-time in surgeries. However, barriers including costs, generalizability, and medicolegal interpretations remain and need to be addressed before the algorithms become common place in clinical practice.
KW - Artificial intelligence
KW - Deep learning
KW - Retina
KW - Retinopathy of prematurity
UR - http://www.scopus.com/inward/record.url?scp=85173905396&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85173905396&partnerID=8YFLogxK
U2 - 10.1007/978-3-031-14506-3_68
DO - 10.1007/978-3-031-14506-3_68
M3 - Chapter
AN - SCOPUS:85173905396
SN - 9783031145056
SP - 1011
EP - 1017
BT - Pediatric Vitreoretinal Surgery
PB - Springer International Publishing
ER -