RIDDLE: Race and ethnicity Imputation from Disease history with Deep LEarning
KAUST DepartmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
Computer Science Program
Computational Bioscience Research Center (CBRC)
Permanent link to this recordhttp://hdl.handle.net/10754/627696
MetadataShow full item record
AbstractAnonymized electronic medical records are an increasingly popular source of research data. However, these datasets often lack race and ethnicity information. This creates problems for researchers modeling human disease, as race and ethnicity are powerful confounders for many health exposures and treatment outcomes; race and ethnicity are closely linked to population-specific genetic variation. We showed that deep neural networks generate more accurate estimates for missing racial and ethnic information than competing methods (e.g., logistic regression, random forest, support vector machines, and gradient-boosted decision trees). RIDDLE yielded significantly better classification performance across all metrics that were considered: accuracy, cross-entropy loss (error), precision, recall, and area under the curve for receiver operating characteristic plots (all p < 10-9). We made specific efforts to interpret the trained neural network models to identify, quantify, and visualize medical features which are predictive of race and ethnicity. We used these characterizations of informative features to perform a systematic comparison of differential disease patterns by race and ethnicity. The fact that clinical histories are informative for imputing race and ethnicity could reflect (1) a skewed distribution of blue- and white-collar professions across racial and ethnic groups, (2) uneven accessibility and subjective importance of prophylactic health, (3) possible variation in lifestyle, such as dietary habits, and (4) differences in background genetic variation which predispose to diseases.
CitationKim J-S, Gao X, Rzhetsky A (2018) RIDDLE: Race and ethnicity Imputation from Disease history with Deep LEarning. PLOS Computational Biology 14: e1006106. Available: http://dx.doi.org/10.1371/journal.pcbi.1006106.
SponsorsThe study was supported by funds from the Defense Advanced Projects Agency, contract W911NF1410333 to AR, https://urldefense.proofpoint.com/v2/url?u=http-3A__www.darpa.mil_program_big-2Dmechanism&d=DwIGaQ&c=Nd1gv_ZWYNIRyZYZmXb18oVfc3lTqv2smA_esABG70U&r=ULvNIgo15mH_8cCBmiM1KBF_qHRW8ZMYO-_ZDPm3uOp9kFqARW63OFcx12Y06DIX&m=t94GIr1nrxziPIHpUNauHQejNovIkVRHPMsNYXkgjNg&s=L72IDKoNZ_89dilSQYA3xsw98W2QbGXv0RdpZxi0oQk&e=, the National Heart Lung and Blood Institute, award R01HL122712 to A.R., https://urldefense.proofpoint.com/v2/url?u=https-3A__www.nhlbi.nih.gov&d=DwIGaQ&c=Nd1gv_ZWYNIRyZYZmXb18oVfc3lTqv2smA_esABG70U&r=ULvNIgo15mH_8cCBmiM1KBF_qHRW8ZMYO-_ZDPm3uOp9kFqARW63OFcx12Y06DIX&m=t94GIr1nrxziPIHpUNauHQejNovIkVRHPMsNYXkgjNg&s=PSG7vwNtEqDmJ-ch-N921YI8xACd-N-EyAJZbHII6Fw&e=, the National Institute of Mental Health, award P50 MH094267 to AR, https://urldefense.proofpoint.com/v2/url?u=https-3A__grants.nih.gov_grants_guide_pa-2Dfiles_PAR-2D14-2D120.html&d=DwIGaQ&c=Nd1gv_ZWYNIRyZYZmXb18oVfc3lTqv2smA_esABG70U&r=ULvNIgo15mH_8cCBmiM1KBF_qHRW8ZMYO-_ZDPm3uOp9kFqARW63OFcx12Y06DIX&m=t94GIr1nrxziPIHpUNauHQejNovIkVRHPMsNYXkgjNg&s=Qu6HXGbIRPOE-yJSf3SFxJxutavu5K_Ic3FkHjau-s0&e=, by the King Abdullah University of Science and Technology (KAUST) Office of Sponsored Research (OSR), awards FCC/1/1976-04, URF/1/3007-01, URF/1/3450-01 and URF/1/3454-01to XG, and a gift from Liz and Kent Dauten to AR. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
PublisherPublic Library of Science (PLoS)
JournalPLOS Computational Biology
Except where otherwise noted, this item's license is described as This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
- Imputing Missing Race/Ethnicity in Pediatric Electronic Health Records: Reducing Bias with Use of U.S. Census Location and Surname Data.
- Authors: Grundmeier RW, Song L, Ramos MJ, Fiks AG, Elliott MN, Fremont A, Pace W, Wasserman RC, Localio R
- Issue date: 2015 Aug
- Concordance between self-reported race/ethnicity and that recorded in a Veteran Affairs electronic medical record.
- Authors: Hamilton NS, Edelman D, Weinberger M, Jackson GL
- Issue date: 2009 Jul-Aug
- Health plan administrative records versus birth certificate records: quality of race and ethnicity information in children.
- Authors: Smith N, Iyer RL, Langer-Gould A, Getahun DT, Strickland D, Jacobsen SJ, Chen W, Derose SF, Koebnick C
- Issue date: 2010 Nov 23
- Assessing and Mitigating Bias in Medical Artificial Intelligence: The Effects of Race and Ethnicity on a Deep Learning Model for ECG Analysis.
- Authors: Noseworthy PA, Attia ZI, Brewer LC, Hayes SN, Yao X, Kapa S, Friedman PA, Lopez-Jimenez F
- Issue date: 2020 Mar
- Missing race/ethnicity data in Veterans Health Administration based disparities research: a systematic review.
- Authors: Long JA, Bamba MI, Ling B, Shea JA
- Issue date: 2006 Feb