Association of Biomarker-Based Artificial Intelligence With Risk of Racial Bias in Retinal Images

Abstract

ImportanceAlthough 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.ObjectiveTo 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 ParticipantsThe 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.Main Outcomes and MeasuresArea 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.ResultsA 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.Conclusions and RelevanceResults 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.

Retina Fundus Photograph-Based Artificial Intelligence Algorithms in Medicine: A Systematic Review

“…Using ML algorithms, comprehensive datasets of health parameters and such associated features as retinal images, electronic health records, and biological markers can be analyzed in an automated, accurate, and efficient manner. For example, AI analysis of retinal images can provide information on age, sex, and race [ 8 – 10 ]. Similarly, AI algorithms trained on electronic health records can detect patterns and predict 47 systemic biomarkers as outcome variables, including BMI, blood pressure, and HbA1c levels [ 10 ].…”

Equitable deep learning for diabetic retinopathy detection using multi-dimensional retinal imaging with fair adaptive scaling: a retrospective study

“…For example, there are fewer Black and Asian DR patients present in ophthalmic care, which is data inequality. 5 In addition, prior studies have shown that retinal anatomy is related to sex and racial information, 18,19 which is an example of data characteristic variability.…”

“…For example, there are fewer Black and Asian DR patients present in ophthalmic care, which is data inequality. 5 In addition, prior studies have shown that retinal anatomy is related to sex and racial information, 18,19 which is an example of data characteristic variability. Mitigating data inequality and addressing data characteristic variability is imperative to reduce performance disparities and achieve more equitable outcomes in deep learning.…”

Deep Transfer Learning for Ethnically Distinct Populations: Prediction of Refractive Error Using Optical Coherence Tomography

“…It has been applied in ophthalmology for identifying vision-related diseases [ 2 , 3 ]. However, despite the known racial and ethnic differences in ocular measurements and pathological conditions [ 4 , 5 ], few studies have examined the application of deep learning in different races. Especially the mismatch between training and testing data distribution causes significant degradation in the model performance in multi-ethnic scenarios.…”