Bias and Non-Diversity of Big Data in Artificial Intelligence: Focus on Retinal Diseases

References

• Bellemo V, Lim G, Rim TH, et al. Artificial intelligence screening for diabetic retinopathy: the real-world emerging application. Curr Diab Rep. 2019;19(9):72. doi:10.1007/s11892-019-1189-3.
   PubMed Web of Science ®Google Scholar
• Abràmoff MD, Lavin PT, Birch M, et al. Pivotal trial of an autonomous AI-based diagnostic system for detection of diabetic retinopathy in primary care offices. NPJ Digital Med. 2018;1(1):1–8. doi:10.1038/s41746-018-0040-6.
   PubMedGoogle Scholar
• Bhaskaranand M, Ramachandra C, Bhat S, et al. The value of automated diabetic retinopathy screening with the EyeArt system: a study of more than 100,000 consecutive encounters from people with diabetes. Diabetes Technol Ther. 2019;21(11):635–643. doi:10.1089/dia.2019.0164.
   PubMed Web of Science ®Google Scholar
• Gianfrancesco MA, Tamang S, Yazdany J, et al. Potential biases in machine learning algorithms using electronic health record data. JAMA Intern Med. 2018;178(11):1544–1547. doi:10.1001/jamainternmed.2018.3763.
   PubMed Web of Science ®Google Scholar
• Agarwal A, Dudík M, Wu ZS Fair regression: quantitative definitions and reduction-based algorithms. Paper presented at: International Conference on Machine Learning, 2019.
   Google Scholar
• Mehrabi N, Morstatter F, Saxena N, et al. A survey on bias and fairness in machine learning. ACM Comput Surveys. 2023;54(6):1–35. doi:10.1145/3457607.
   Web of Science ®Google Scholar
• Aghaei S, Azizi MJ, Vayanos P Learning optimal and fair decision trees for non-discriminative decision-making. Paper presented at: Proceedings of the AAAI Conference on Artificial Intelligence, 2019.
   Google Scholar
• Alipourfard N, Fennell PG, Lerman K Can you Trust the Trend? Discovering Simpson’s Paradoxes in Social Data. Paper presented at: Proceedings of the Eleventh ACM International Conference on Web Search and Data Mining, 2018.
   Google Scholar
• Jacoba CMP, Celi LA, Silva PS. Biomarkers for Progression in Diabetic Retinopathy: expanding Personalized Medicine through Integration of AI with Electronic Health Records. Semin Ophthalmol. 2021;36(4):250–257. doi:10.1080/08820538.2021.1893351.
   PubMed Web of Science ®Google Scholar
• Ng JH, Ye F, Ward LM, et al. Data on race, ethnicity, and language largely incomplete for managed care plan members. Health Affairs. 2017;36(3):548–552. doi:10.1377/hlthaff.2016.1044.
   Web of Science ®Google Scholar
• Arpey NC, Gaglioti AH, Rosenbaum ME. How socioeconomic status affects patient perceptions of health care: a qualitative study. J Prim Care Community Health. 2017;8(3):169–175. doi:10.1177/2150131917697439.
   PubMed Web of Science ®Google Scholar
• Ramoni M, Sebastiani P. Robust learning with missing data. Mach Learn. 2001;45(2):147–170. doi:10.1023/A:1010968702992.
   Web of Science ®Google Scholar
• Liu Y, Zupan NJ, Swearingen R, et al. Identification of barriers, facilitators and system-based implementation strategies to increase teleophthalmology use for diabetic eye screening in a rural US primary care clinic: a qualitative study. BMJ open. 2019;9(2):e022594. doi:10.1136/bmjopen-2018-022594.
   PubMed Web of Science ®Google Scholar
• Rauscher GH, Khan JA, Berbaum ML, et al. Potentially missed detection with screening mammography: does the quality of radiologist’s interpretation vary by patient socioeconomic advantage/disadvantage? Ann Epidemiol. 2013;23(4):210–214. doi:10.1016/j.annepidem.2013.01.006.
   PubMed Web of Science ®Google Scholar
• Khan SM, Liu X, Nath S, et al. A global review of publicly available datasets for ophthalmological imaging: barriers to access, usability, and generalisability. The Lancet Digital Health. 2021;3(1):e51–e66. doi:10.1016/S2589-7500(20)30240-5.
   PubMedGoogle Scholar
• Kim DW, Jang HY, Kim KW, et al. Design characteristics of studies reporting the performance of artificial intelligence algorithms for diagnostic analysis of medical images: results from recently published papers. Korean journal of radiology. 2019;20(3):405–410.
   Google Scholar
• Kanagasingam Y, Xiao D, Vignarajan J, et al. Evaluation of artificial intelligence–based grading of diabetic retinopathy in primary care. JAMA network open. e182665-e182665. 2018;1(5):e182665. doi:10.1001/jamanetworkopen.2018.2665.
   PubMed Web of Science ®Google Scholar
• Gulshan V, Rajan RP, Widner K, et al. Performance of a deep-learning algorithm vs manual grading for detecting diabetic retinopathy in India. JAMA Ophthalmol. 2019;137(9):987–993. doi:10.1001/jamaophthalmol.2019.2004.
   PubMed Web of Science ®Google Scholar
• Ibrahim H, Liu X, Rivera SC, et al. Reporting guidelines for clinical trials of artificial intelligence interventions: the SPIRIT-AI and CONSORT-AI guidelines. Trials. 2021;22(1):1–5. doi:10.1186/s13063-020-04951-6.
   PubMed Web of Science ®Google Scholar
• Sounderajah V, Ashrafian H, Aggarwal R, et al. Developing specific reporting guidelines for diagnostic accuracy studies assessing AI interventions: the STARD-AI Steering Group. Nat Med. 2020;26(6):807–808. doi:10.1038/s41591-020-0941-1.
   PubMed Web of Science ®Google Scholar
• Collins GS, Moons KG. Reporting of artificial intelligence prediction models. Lancet. 2019;393(10181):1577–1579. doi:10.1016/S0140-6736(19)30037-6.
   PubMed Web of Science ®Google Scholar
• Wawira Gichoya J, McCoy LG, Celi LAG, et al. Equity in essence: a call for operationalising fairness in machine learning for healthcare. 2021.
   Google Scholar
• Yiu G, Vuong VS, Oltjen S, et al. Effect of uveal melanocytes on choroidal morphology in rhesus macaques and humans on enhanced-depth imaging optical coherence tomography. Invest Ophthalmol Vis Sci. 2016;57(13):5764–5771. doi:10.1167/iovs.16-20070.
   PubMed Web of Science ®Google Scholar
• Li X, Wong WL, Cheung CY-L, et al. Racial differences in retinal vessel geometric characteristics: a multiethnic study in healthy Asians. Invest Ophthalmol Vis Sci. 2013;54(5):3650–3656. doi:10.1167/iovs.12-11126.
   PubMed Web of Science ®Google Scholar
• Yau JWY, Xie J, Lamoureux E, et al. Retinal microvascular calibre and risk of incident diabetes: the multi-ethnic study of atherosclerosis. Diabetes Res Clin Pract. 2012;95(2):265–274. doi:10.1016/j.diabres.2011.10.027.
   PubMed Web of Science ®Google Scholar
• Wong TY, Islam FA, Klein R, et al. Retinal vascular caliber, cardiovascular risk factors, and inflammation: the multi-ethnic study of atherosclerosis (Mesa). Invest Ophthalmol Vis Sci. 2006;47(6):2341–2350. doi:10.1167/iovs.05-1539.
   PubMed Web of Science ®Google Scholar
• Klein R, Klein BE, Moss SE, et al. Retinal vessel caliber and microvascular and macrovascular disease in type 2 diabetes: XXI: the Wisconsin Epidemiologic Study of Diabetic Retinopathy. Ophthalmology. 2007;114(10):1884–1892. doi:10.1016/j.ophtha.2007.02.023.
   PubMed Web of Science ®Google Scholar
• Kawasaki R, Cheung N, Wang JJ, et al. Retinal vessel diameters and risk of hypertension: the Multiethnic Study of Atherosclerosis. J Hypertens. 2009;27(12):2386.
   PubMed Web of Science ®Google Scholar
• Ikram MK, Witteman JC, Vingerling JR, et al. Retinal vessel diameters and risk of hypertension: the Rotterdam Study. hypertension. 2006;47(2):189–194. doi:10.1161/01.HYP.0000199104.61945.33.
   PubMed Web of Science ®Google Scholar
• Joachim N, Mitchell P, Younan C, et al. Ethnic variation in early age-related macular degeneration lesions between white Australians and Singaporean Asians. Invest Ophthalmol Vis Sci. 2014;55(7):4421–4429. doi:10.1167/iovs.14-14476.
   PubMed Web of Science ®Google Scholar
• Green WR, Enger C. Age-related macular degeneration histopathologic studies: the 1992 Lorenz E. Zimmerman Lecture. Ophthalmology. 1993;100(10):1519–1535. doi:10.1016/S0161-6420(93)31466-1.
   PubMed Web of Science ®Google Scholar
• Jampol LM, Tielsch J. Race, macular degeneration, and the Macular Photocoagulation Study. Arch Ophthalmol. 1992;110(12):1699–1700. doi:10.1001/archopht.1992.01080240039024.
   PubMedGoogle Scholar
• d’Alessandro B, O’Neil C, LaGatta T. Conscientious classification: a data scientist’s guide to discrimination-aware classification. Big Data. 2017;5(2):120–134. doi:10.1089/big.2016.0048.
   PubMed Web of Science ®Google Scholar
• Joshi N, Burlina P. AI Fairness via Domain Adaptation. arXiv preprint arXiv:2104.01109. 2021. https://arxiv.org/pdf/2104.01109.pdf
   Google Scholar
• Burlina P, Joshi N, Paul W, et al. Addressing artificial intelligence bias in retinal disease diagnostics. Translational Vision Science & Technology, 2021; 10(2): 13–13. doi:10.1167/tvst.10.2.13
   Google Scholar
• Katada Y, Ozawa N, Masayoshi K, et al. Automatic screening for diabetic retinopathy in interracial fundus images using artificial intelligence. Intellig Med. 2020;3:100024. doi:10.1016/j.ibmed.2020.100024.
   Google Scholar
• Burlina P, Paul W, Mathew P, et al. Low-shot deep learning of diabetic retinopathy with potential applications to address artificial intelligence bias in retinal diagnostics and rare ophthalmic diseases. JAMA Ophthalmol. 2020;138(10):1070–1077. doi:10.1001/jamaophthalmol.2020.3269.
   PubMed Web of Science ®Google Scholar
• Marmot M, Friel S, Bell R, et al. Closing the gap in a generation: health equity through action on the social determinants of health. Lancet. 2008;372(9650):1661–1669. doi:10.1016/S0140-6736(08)61690-6.
   PubMed Web of Science ®Google Scholar
• Zhao Y, Wood EP, Mirin N, et al. Social determinants in machine learning cardiovascular disease prediction models: a systematic review. Am J Prev Med. 2021;61(4):596–605. doi:10.1016/j.amepre.2021.04.016.
   PubMed Web of Science ®Google Scholar
• Alaa AM, Bolton T, Di Angelantonio E, et al. Cardiovascular disease risk prediction using automated machine learning: a prospective study of 423,604 UK Biobank participants. PloS one. 2019;14(5):e0213653. doi:10.1371/journal.pone.0213653.
   PubMed Web of Science ®Google Scholar
• Zhang X, Cotch MF, Ryskulova A, et al. Vision health disparities in the United States by race/ethnicity, education, and economic status: findings from two nationally representative surveys. Am J Ophthalmol. e51. 2012;154(6):S53–S62. doi:10.1016/j.ajo.2011.08.045.
   PubMed Web of Science ®Google Scholar
• Silverberg EL, Sterling TW, Williams TH, Castro G, Rodriguez de la Vega P, Barengo NC, Silverberg EL, Sterling TW, Williams TH, et al. The association between social determinants of health and self-reported diabetic retinopathy: an exploratory analysis. Int J Environ Res Public Health. 2021;18(2):792. doi:10.3390/ijerph18020792.
   PubMed Web of Science ®Google Scholar
• Gomes MB, Santos DC, Pizarro MH, et al. Relationship between health care insurance status, social determinants and prevalence of diabetes-related microvascular complications in patients with Type 1 diabetes: a nationwide survey in Brazil. Acta Diabetol. 2019;56(6):697–705. doi:10.1007/s00592-019-01308-7.
   PubMed Web of Science ®Google Scholar
• Silva PS, Horton MB, Clary D, et al. Identification of diabetic retinopathy and ungradable image rate with ultrawide field imaging in a national teleophthalmology program. Ophthalmology. 2016;123(6):1360–1367. doi:10.1016/j.ophtha.2016.01.043.
   PubMed Web of Science ®Google Scholar
• American Diabetes Association. Economic costs of diabetes in the US in 2017. Diabetes Care. 2018;41(5):917–928. doi:10.2337/dci18-0007.
   PubMed Web of Science ®Google Scholar
• Orton E, Forbes-Haley A, Tunbridge L, et al. Equity of uptake of a diabetic retinopathy screening programme in a geographically and socio-economically diverse population. Public Health. 2013;127(9):814–821. doi:10.1016/j.puhe.2013.04.015.
   PubMed Web of Science ®Google Scholar
• Estevez JJ, Howard NJ, Craig JE, et al. Working towards eye health equity for indigenous Australians with diabetes. Int J Environ Res Public Health. 2019;16(24):5060. doi:10.3390/ijerph16245060.
   PubMed Web of Science ®Google Scholar
• Malhotra NA, Hom GL, Conti T, et al. Characterizing how racial and socioeconomic factors affect anti-VEGF treatment utilization and outcomes for diabetic macular edema. Invest Ophthalmol Vis Sci. 2020;61(7):3292.
   Web of Science ®Google Scholar
• Kras A, Celi LA, Miller JB. Accelerating ophthalmic artificial intelligence research: the role of an open access data repository. Curr Opin Ophthalmol. 2020;31(5):337–350. doi:10.1097/ICU.0000000000000678.
   PubMed Web of Science ®Google Scholar
• National Institutes of Health. Bridge to AI Program (Bridge2AI). https://commonfund.nih.gov/bridge2ai. Accessed December 22, 2021.
   Google Scholar
• Yang Q, Liu Y, Chen T, et al. Federated machine learning: concept and applications. ACM Transac Intellig Syst Technol. 2019;10(2):1–19. doi:10.1145/3339474.
   Web of Science ®Google Scholar
• Gebru T, Morgenstern J, Vecchione B, et al. Datasheets for datasets. In Ethics of Data and Analytics. Auerbach Publications; 2022: 148–156.
   Google Scholar
• Barabas C, Virza M, Dinakar K, et al. Interventions over predictions: reframing the ethical debate for actuarial risk assessment. Paper presented at: Conference on Fairness, Accountability and Transparency. New York University, New York City, 2018.
   Google Scholar
• Barocas S, Selbst AD. Big data’s disparate impact. Calif L Rev. 2016;104:671.
   Web of Science ®Google Scholar
• Hall M, Frank E, Holmes G, et al. The WEKA data mining software: an update. ACM SIGKDD Explorations Newsl. 2009;11(1):10–18. doi:10.1145/1656274.1656278.
   Google Scholar
• Faes L, Wagner SK, Fu DJ, et al. Automated deep learning design for medical image classification by health-care professionals with no coding experience: a feasibility study. The Lancet Digital Health. e232-e242. 2019;1(5):e232–e242. doi:10.1016/S2589-7500(19)30108-6.
   PubMedGoogle Scholar
• Kim IK, Lee K, Park JH, et al. Classification of pachychoroid disease on ultrawide-field indocyanine green angiography using auto-machine learning platform. British J Ophthalmol. 2021;105(6):856–861. doi:10.1136/bjophthalmol-2020-316108.
   PubMed Web of Science ®Google Scholar
• Korot E, Pontikos N, Liu X, et al. Predicting sex from retinal fundus photographs using automated deep learning. Sci Rep. 2021;11(1):1–8. doi:10.1038/s41598-021-89743-x.
   PubMed Web of Science ®Google Scholar
• Antaki F, Coussa RG, Kahwati G, et al. Accuracy of automated machine learning in classifying retinal pathologies from ultra-widefield pseudocolour fundus images. British J Ophthalmol. 2021;107(1):90–95. doi:10.1136/bjophthalmol-2021-319030.
   PubMed Web of Science ®Google Scholar
• Caton S, Haas C. Fairness in machine learning: a survey. arXiv preprint arXiv:2010.04053. 2020. https://arxiv.org/pdf/2010.04053.pdf
   Google Scholar
• Hardt M, Price E, Srebro N. Equality of opportunity in supervised learning. Adv Neural Inform Process Sys. 2016;29:3315–3323.
   Google Scholar
• Zliobaite I. On the relation between accuracy and fairness in binary classification. arXiv preprint arXiv:1505.05723. 2015. https://arxiv.org/pdf/1505.05723.pdf
   Google Scholar
• Corbett-Davies S, Pierson E, Feller A, et al. Algorithmic decision making and the cost of fairness. Paper presented at: Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Halifax, Nova Scotia, Canada, 2017.
   Google Scholar
• Speicher T, Heidari H, Grgic-Hlaca N, et al. A unified approach to quantifying algorithmic unfairness: measuring individual &group unfairness via inequality indices. Paper presented at: Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. London, United Kingdom, 2018.
   Google Scholar
• Kleinberg J, Ludwig J, Mullainathan S, et al. Algorithmic fairness. Paper presented at: Aea papers and proceedings, 2018.
   Google Scholar