Secondary Analysis of the Data from the Telemedicine Approaches to Evaluating Acute-phase ROP (e-ROP)

 描述（由申请人提供）：拟议项目将利用远程医疗方法评估早产儿急性期视网膜病变（e-ROP）的综合数据集，这是一项由国家眼科研究所赞助的大规模多中心研究，旨在评估远程医疗系统识别患有严重ROP需要眼科医生评估的婴儿的有效性。这项建议的具体目的并不包括电子人事登记研究的主要目的，因为这些目的已经子索。相反，拟议项目的目标涉及对e-ROP数据进行额外的二次分析，以进一步评估远程医疗对ROP的有效性，并完善图像评估系统，以优化准确性和效率。实现这些目标的具体目的是：（1）描述图像评估结果的特征，包括婴儿双眼之间ROP特征（阶段、区域和加疾病）的对称性、首次检测的时间以及ROP特征随时间的进展。(2)描述通过影像学评价比通过诊断检查更早发现RW-ROP的眼睛的特征。（3）评价是否可以基于使用影像学评价和成像时的体重增加评估的组合的风险分层来减少ROP评价的数量。(4)旨在评价单个读片员系统是否提供与具有两名读片员的系统相同的灵敏度和特异性，并由第三名读片员裁定差异，并描述需要裁定的图像特征。(5)将临床诊断检查结果视为不完善的潜在类别分析，计算图像评价的敏感性和特异性。e-ROP数据集是实现这些目标的独特而宝贵的资源。对1257名婴儿的7749个图像集的图像分级结果进行分析，可能会导致制定更准确和有效的分级方案，用于识别眼科医生进行ROP评估的高危婴儿。非常大的样本量，高质量的数据集，以及丰富的经验和高生产力的研究人员沿着的全力支持下，e-ROP合作小组，支持的可行性和可能性，高产量的拟议项目。

 DESCRIPTION (provided by applicant): The proposed project will utilize the comprehensive dataset from the Telemedicine Approaches to Evaluating Acute-Phase Retinopathy of Prematurity (e-ROP), a large-scale multi-center study, sponsored by the National Eye Institute, to evaluate the validity of a telemedicine system to identify infants who have sufficiently severe ROP to require evaluation by an ophthalmologist. The specific aims of this proposal do not include the major goals of the e-ROP Study, which have been accomplished already. Instead, the goals of the proposed project involve additional secondary analyses of the e-ROP data to further evaluate the validity of telemedicine for ROP and to refine the system for image evaluation to optimize accuracy and efficiency. The Specific Aims to achieve these goals are: (1) To describe characteristics of image evaluation findings, including the symmetry of ROP features (stage, zone, and plus disease) between paired eyes of infants, the time of their first detection, and the progression of ROP features over time. (2) To describe the characteristics of eyes that were found to have RW-ROP by image evaluation earlier than by diagnostic exam. (3) To evaluate whether the number of ROP evaluations can be reduced based on the risk stratification using a combination of image evaluation and weight gain assessment at the time of imaging. (4) To evaluate whether a single reader system provides equivalent sensitivity and specificity as a system with two readers and adjudication of discrepancies by a third reader, and to describe the characteristics of images that require adjudication. (5) To calculate the sensitiviy and specificity of image evaluation using latent class analysis that considers the clinical diagnostic examination finding as imperfect. The e-ROP dataset is a unique and valuable resource for accomplishing these Aims. The analyses of the image grading results from 7749 image sets of 1257 infants may lead to the development of more accurate and efficient grading protocol for identifying high risk infants for ROP evaluation by an ophthalmologist. The very large sample size, high quality of the dataset, as well as the rich experience and high productivity of the investigators along with the full support of the e-ROP Collaborative Group, support the feasibility and likelihood of high yield of the proposed project.