Department of Ophthalmology, Casey Eye Institute, Oregon Health & Science University, Portland.
Radiology, MGH/Harvard Medical School, Charlestown, Massachusetts.
JAMA Ophthalmol. 2023 Jun 1;141(6):543-552. doi: 10.1001/jamaophthalmol.2023.1310.
Although race is a social construct, it is associated with variations in skin and retinal pigmentation. Image-based medical artificial intelligence (AI) algorithms that use images of these organs have the potential to learn features associated with self-reported race (SRR), which increases the risk of racially biased performance in diagnostic tasks; understanding whether this information can be removed, without affecting the performance of AI algorithms, is critical in reducing the risk of racial bias in medical AI.
To evaluate whether converting color fundus photographs to retinal vessel maps (RVMs) of infants screened for retinopathy of prematurity (ROP) removes the risk for racial bias.
DESIGN, SETTING, AND PARTICIPANTS: The retinal fundus images (RFIs) of neonates with parent-reported Black or White race were collected for this study. A u-net, a convolutional neural network (CNN) that provides precise segmentation for biomedical images, was used to segment the major arteries and veins in RFIs into grayscale RVMs, which were subsequently thresholded, binarized, and/or skeletonized. CNNs were trained with patients' SRR labels on color RFIs, raw RVMs, and thresholded, binarized, or skeletonized RVMs. Study data were analyzed from July 1 to September 28, 2021.
Area under the precision-recall curve (AUC-PR) and area under the receiver operating characteristic curve (AUROC) at both the image and eye level for classification of SRR.
A total of 4095 RFIs were collected from 245 neonates with parent-reported Black (94 [38.4%]; mean [SD] age, 27.2 [2.3] weeks; 55 majority sex [58.5%]) or White (151 [61.6%]; mean [SD] age, 27.6 [2.3] weeks, 80 majority sex [53.0%]) race. CNNs inferred SRR from RFIs nearly perfectly (image-level AUC-PR, 0.999; 95% CI, 0.999-1.000; infant-level AUC-PR, 1.000; 95% CI, 0.999-1.000). Raw RVMs were nearly as informative as color RFIs (image-level AUC-PR, 0.938; 95% CI, 0.926-0.950; infant-level AUC-PR, 0.995; 95% CI, 0.992-0.998). Ultimately, CNNs were able to learn whether RFIs or RVMs were from Black or White infants regardless of whether images contained color, vessel segmentation brightness differences were nullified, or vessel segmentation widths were uniform.
Results of this diagnostic study suggest that it can be very challenging to remove information relevant to SRR from fundus photographs. As a result, AI algorithms trained on fundus photographs have the potential for biased performance in practice, even if based on biomarkers rather than raw images. Regardless of the methodology used for training AI, evaluating performance in relevant subpopulations is critical.
尽管种族是一种社会建构，但它与皮肤和视网膜色素沉着的变化有关。使用这些器官图像的基于图像的医学人工智能 (AI) 算法有可能学习与自我报告种族 (SRR) 相关的特征，这增加了在诊断任务中出现种族偏见的风险；了解是否可以去除这些信息，而不影响 AI 算法的性能，对于降低医疗 AI 中的种族偏见风险至关重要。
评估对早产儿视网膜病变 (ROP) 筛查的新生儿的眼底彩色照片进行视网膜血管图 (RVM) 转换是否可以消除种族偏见的风险。
设计、设置和参与者：本研究收集了具有父母报告的黑种人或白种人种族的新生儿的视网膜眼底图像 (RFIs)。使用 U-Net（一种为生物医学图像提供精确分割的卷积神经网络 (CNN)）将 RFIs 中的主要动静脉分割成灰度 RVM，随后对 RVM 进行阈值处理、二值化和/或骨架化。在彩色 RFIs、原始 RVM、经阈值处理、二值化或骨架化的 RVM 上，使用患者的 SRR 标签对 CNN 进行训练。研究数据于 2021 年 7 月 1 日至 9 月 28 日进行分析。
用于 SRR 分类的图像和眼部水平的精度-召回曲线下面积 (AUC-PR) 和接收者操作特征曲线下面积 (AUROC)。
从 245 名父母报告为黑种人 (94 [38.4%]；平均 [标准差] 年龄，27.2 [2.3] 周；55 名多数为男性 [58.5%]) 或白种人 (151 [61.6%]；平均 [标准差] 年龄，27.6 [2.3] 周，80 名多数为男性 [53.0%]) 的新生儿中收集了 4095 张 RFIs。CNN 几乎可以从 RFIs 中推断出 SRR（图像水平 AUC-PR，0.999；95%CI，0.999-1.000；婴儿水平 AUC-PR，1.000；95%CI，0.999-1.000）。原始 RVM 与彩色 RFIs 一样具有信息性（图像水平 AUC-PR，0.938；95%CI，0.926-0.950；婴儿水平 AUC-PR，0.995；95%CI，0.992-0.998）。最终，无论图像是否包含颜色、血管分割亮度差异是否为零，或者血管分割宽度是否均匀，CNN 都能够学习 RFIs 或 RVMs 是来自黑种人还是白种人婴儿。
这项诊断研究的结果表明，从眼底照片中去除与 SRR 相关的信息可能极具挑战性。因此，即使基于生物标志物而不是原始图像，基于眼底照片训练的 AI 算法在实践中也有可能表现出有偏见的性能。无论 AI 采用何种方法进行训练，在相关亚人群中评估性能都至关重要。
