Applied Physics Laboratory, Johns Hopkins University, Laurel, Maryland.
Department of Computer Science and Malone Center for Engineering in Healthcare, Johns Hopkins University, Baltimore, Maryland.
JAMA Ophthalmol. 2023 Jul 1;141(7):677-685. doi: 10.1001/jamaophthalmol.2023.2271.
Best-corrected visual acuity (BCVA) is a measure used to manage diabetic macular edema (DME), sometimes suggesting development of DME or consideration of initiating, repeating, withholding, or resuming treatment with anti-vascular endothelial growth factor. Using artificial intelligence (AI) to estimate BCVA from fundus images could help clinicians manage DME by reducing the personnel needed for refraction, the time presently required for assessing BCVA, or even the number of office visits if imaged remotely.
To evaluate the potential application of AI techniques for estimating BCVA from fundus photographs with and without ancillary information.
DESIGN, SETTING, AND PARTICIPANTS: Deidentified color fundus images taken after dilation were used post hoc to train AI systems to perform regression from image to BCVA and to evaluate resultant estimation errors. Participants were patients enrolled in the VISTA randomized clinical trial through 148 weeks wherein the study eye was treated with aflibercept or laser. The data from study participants included macular images, clinical information, and BCVA scores by trained examiners following protocol refraction and VA measurement on Early Treatment Diabetic Retinopathy Study (ETDRS) charts.
Primary outcome was regression evaluated by mean absolute error (MAE); the secondary outcome included percentage of predictions within 10 letters, computed over the entire cohort as well as over subsets categorized by baseline BCVA, determined from baseline through the 148-week visit.
Analysis included 7185 macular color fundus images of the study and fellow eyes from 459 participants. Overall, the mean (SD) age was 62.2 (9.8) years, and 250 (54.5%) were male. The baseline BCVA score for the study eyes ranged from 73 to 24 letters (approximate Snellen equivalent 20/40 to 20/320). Using ResNet50 architecture, the MAE for the testing set (n = 641 images) was 9.66 (95% CI, 9.05-10.28); 33% of the values (95% CI, 30%-37%) were within 0 to 5 letters and 28% (95% CI, 25%-32%) within 6 to 10 letters. For BCVA of 100 letters or less but more than 80 letters (20/10 to 20/25, n = 161) and 80 letters or less but more than 55 letters (20/32 to 20/80, n = 309), the MAE was 8.84 letters (95% CI, 7.88-9.81) and 7.91 letters (95% CI, 7.28-8.53), respectively.
This investigation suggests AI can estimate BCVA directly from fundus photographs in patients with DME, without refraction or subjective visual acuity measurements, often within 1 to 2 lines on an ETDRS chart, supporting this AI concept if additional improvements in estimates can be achieved.
最佳矫正视力（BCVA）是用于管理糖尿病性黄斑水肿（DME）的一种测量方法，有时可以提示 DME 的发展，或者考虑开始、重复、停止或恢复使用抗血管内皮生长因子治疗。使用人工智能（AI）从眼底图像估计 BCVA 可以帮助临床医生通过减少对折射的人员需求，减少目前评估 BCVA 所需的时间，甚至如果远程成像的话，减少就诊次数来管理 DME。
评估 AI 技术在有和没有辅助信息的情况下从眼底照片中估计 BCVA 的潜在应用。
设计、地点和参与者：使用事后获得的散瞳后的彩色眼底图像，训练 AI 系统从图像到 BCVA 进行回归，并评估由此产生的估计误差。参与者是通过 148 周的 VISTA 随机临床试验招募的患者，其中研究眼接受了阿柏西普或激光治疗。研究参与者的数据包括黄斑图像、临床信息和经过训练的检查者根据协议进行折射和视力测量后获得的 BCVA 分数，视力测量是在早期糖尿病性视网膜病变研究（ETDRS）图表上进行的。
主要结果是通过平均绝对误差（MAE）进行回归评估；次要结果包括在整个队列中以及在按基线 BCVA 分类的亚组（从基线到 148 周就诊）中，计算出的 10 个字母内预测的百分比。
分析包括 459 名参与者的 7185 个研究眼和对侧眼的黄斑彩色眼底图像。总体而言，平均（SD）年龄为 62.2（9.8）岁，250 名（54.5%）为男性。研究眼的基线 BCVA 评分范围为 73 至 24 个字母（约相当于 Snellen 等效 20/40 至 20/320）。使用 ResNet50 架构，测试集（n=641 张图像）的 MAE 为 9.66（95%CI，9.05-10.28）；33%（95%CI，30%-37%）的值在 0 到 5 个字母之间，28%（95%CI，25%-32%）在 6 到 10 个字母之间。对于 BCVA 为 100 个字母或更少但大于 80 个字母（20/10 至 20/25，n=161）和 80 个字母或更少但大于 55 个字母（20/32 至 20/80，n=309），MAE 分别为 8.84 个字母（95%CI，7.88-9.81）和 7.91 个字母（95%CI，7.28-8.53）。
这项研究表明，人工智能可以直接从 DME 患者的眼底照片中估计 BCVA，无需进行折射或主观视力测量，通常在 ETDRS 图表上的 1 到 2 行内，支持这一人工智能概念，如果可以进一步提高估计的准确性。
