Search

Background/objectives: With the increasing survival of premature infants, there is an increased demand to provide adequate retinopathy of prematurity (ROP) services. Wide field retinal imaging (WFDRI) and artificial intelligence (AI) have shown promise in the field of ROP and have the potential to improve the diagnostic performance and reduce the workload for screening ophthalmologists. The aim of this review is to systematically review and provide a summary of the diagnostic characteristics of existing deep learning algorithms., Subject/methods: Two authors independently searched the literature, and studies using a deep learning system from retinal imaging were included. Data were extracted, assessed and reported using PRISMA guidelines., Results: Twenty-seven studies were included in this review. Nineteen studies used AI systems to diagnose ROP, classify the staging of ROP, diagnose the presence of pre-plus or plus disease, or assess the quality of retinal images. The included studies reported a sensitivity of 71%-100%, specificity of 74-99% and area under the curve of 91-99% for the primary outcome of the study. AI techniques were comparable to the assessment of ophthalmologists in terms of overall accuracy and sensitivity. Eight studies evaluated vascular severity scores and were able to accurately differentiate severity using an automated classification score., Conclusion: Artificial intelligence for ROP diagnosis is a growing field, and many potential utilities have already been identified, including the presence of plus disease, staging of disease and a new automated severity score. AI has a role as an adjunct to clinical assessment; however, there is insufficient evidence to support its use as a sole diagnostic tool currently., (© 2022. Crown.)

Introduction: Retinopathy of prematurity (ROP) is the most common disease leading to blindness in extreme preterm infants. Current screening guidelines recommend frequent eye examinations. There is a dearth of trained ophthalmologists for these frequent screening procedures. The ANZNN neonatal network report (2013) found that only 6.4% of all screened infants had severe ROP and less than half received treatment. WINROP (online prediction model, Sweden) uses the postnatal weight gain (surrogate marker for low insulin-like growth factor IGF-1 and poor retinal vascular growth) to identify ROP requiring treatment and aims to reduce the number of examinations. Our objective was to validate the WINROP model in an Australian cohort of preterm infants., Methods: Birth weight, gestational age, and weekly weight measurements were retrieved retrospectively along with the final ROP outcomes and plotted on the online WINROP software., Results: The sensitivity, specificity, positive predictive value, and negative predictive value of WINROP were 85.7%, 59.0%, 6.98%, and 99.1% respectively for a cohort of 221 preterm infants (Median birth weight, 1040 g; Gestational age, 27.9 weeks). WINROP alarm was signaled in 42.6% of all infants. WINROP did not signal an alarm in one infant who needed treatment. This infant had intra ventricular hemorrhage grade 3-4 and temporary ventricular dilatation., Conclusions: This is the first Australian study validating WINROP model. Our findings suggest that it lacked sensitivity to be used alone. However, adjusting the algorithm for the Australian population may improve the efficacy and reduce the number of examinations when used along with the current screening guidelines.