Campbell J Peter, Singh Praveer, Redd Travis K, Brown James M, Shah Parag K, Subramanian Prema, Rajan Renu, Valikodath Nita, Cole Emily, Ostmo Susan, Chan R V Paul, Venkatapathy Narendran, Chiang Michael F, Kalpathy-Cramer Jayashree
Department of Ophthalmology, Casey Eye Institute and
Contributed equally as co-first authors.
Pediatrics. 2021 Mar;147(3). doi: 10.1542/peds.2020-016618.
Childhood blindness from retinopathy of prematurity (ROP) is increasing as a result of improvements in neonatal care worldwide. We evaluate the effectiveness of artificial intelligence (AI)-based screening in an Indian ROP telemedicine program and whether differences in ROP severity between neonatal care units (NCUs) identified by using AI are related to differences in oxygen-titrating capability.
External validation study of an existing AI-based quantitative severity scale for ROP on a data set of images from the Retinopathy of Prematurity Eradication Save Our Sight ROP telemedicine program in India. All images were assigned an ROP severity score (1-9) by using the Imaging and Informatics in Retinopathy of Prematurity Deep Learning system. We calculated the area under the receiver operating characteristic curve and sensitivity and specificity for treatment-requiring retinopathy of prematurity. Using multivariable linear regression, we evaluated the mean and median ROP severity in each NCU as a function of mean birth weight, gestational age, and the presence of oxygen blenders and pulse oxygenation monitors.
The area under the receiver operating characteristic curve for detection of treatment-requiring retinopathy of prematurity was 0.98, with 100% sensitivity and 78% specificity. We found higher median (interquartile range) ROP severity in NCUs without oxygen blenders and pulse oxygenation monitors, most apparent in bigger infants (>1500 g and 31 weeks' gestation: 2.7 [2.5-3.0] vs 3.1 [2.4-3.8]; = .007, with adjustment for birth weight and gestational age).
Integration of AI into ROP screening programs may lead to improved access to care for secondary prevention of ROP and may facilitate assessment of disease epidemiology and NCU resources.
由于全球新生儿护理水平的提高，早产儿视网膜病变（ROP）导致的儿童失明现象正在增加。我们评估了基于人工智能（AI）的筛查在印度ROP远程医疗项目中的有效性，以及使用AI识别的新生儿护理单元（NCU）之间ROP严重程度的差异是否与氧滴定能力的差异有关。
对印度早产儿视网膜病变消除拯救我们的视力ROP远程医疗项目图像数据集上现有的基于AI的ROP定量严重程度量表进行外部验证研究。使用早产儿视网膜病变深度学习系统中的成像和信息学，为所有图像分配一个ROP严重程度评分（1 - 9）。我们计算了治疗所需早产儿视网膜病变的受试者操作特征曲线下面积、敏感性和特异性。使用多变量线性回归，我们评估了每个NCU中ROP严重程度的均值和中位数，作为平均出生体重、胎龄以及是否存在氧气混合器和脉搏氧饱和度监测仪的函数。
检测治疗所需早产儿视网膜病变的受试者操作特征曲线下面积为0.98，敏感性为100%，特异性为78%。我们发现，在没有氧气混合器和脉搏氧饱和度监测仪的NCU中，ROP严重程度的中位数（四分位间距）更高，在较大婴儿（>1500 g且孕31周：2.7 [2.5 - 3.0] 对 3.1 [2.4 - 3.8]；P = .007，校正出生体重和胎龄后）中最为明显。
将AI整合到ROP筛查项目中可能会改善ROP二级预防的医疗服务可及性，并有助于评估疾病流行病学和NCU资源。
