Confounders mediate AI prediction of demographics in medical imaging
Deep learning has been shown to accurately assess “hidden” phenotypes from medical imaging beyond traditional clinician interpretation. Using large echocardiography datasets from two healthcare systems, we test whether it is possible to predict age, race, and sex from cardiac ultrasound images using...
Saved in:
Published in | NPJ digital medicine Vol. 5; no. 1; p. 188 |
---|---|
Main Authors | , , , , , , |
Format | Journal Article |
Language | English |
Published |
London
Nature Publishing Group UK
22.12.2022
Nature Publishing Group Nature Portfolio |
Subjects | |
Online Access | Get full text |
Cover
Loading…
Abstract | Deep learning has been shown to accurately assess “hidden” phenotypes from medical imaging beyond traditional clinician interpretation. Using large echocardiography datasets from two healthcare systems, we test whether it is possible to predict age, race, and sex from cardiac ultrasound images using deep learning algorithms and assess the impact of varying confounding variables. Using a total of 433,469 videos from Cedars-Sinai Medical Center and 99,909 videos from Stanford Medical Center, we trained video-based convolutional neural networks to predict age, sex, and race. We found that deep learning models were able to identify age and sex, while unable to reliably predict race. Without considering confounding differences between categories, the AI model predicted sex with an AUC of 0.85 (95% CI 0.84–0.86), age with a mean absolute error of 9.12 years (95% CI 9.00–9.25), and race with AUCs ranging from 0.63 to 0.71. When predicting race, we show that tuning the proportion of confounding variables (age or sex) in the training data significantly impacts model AUC (ranging from 0.53 to 0.85), while sex and age prediction was not particularly impacted by adjusting race proportion in the training dataset AUC of 0.81–0.83 and 0.80–0.84, respectively. This suggests significant proportion of AI’s performance on predicting race could come from confounding features being detected. Further work remains to identify the particular imaging features that associate with demographic information and to better understand the risks of demographic identification in medical AI as it pertains to potentially perpetuating bias and disparities. |
---|---|
AbstractList | Deep learning has been shown to accurately assess "hidden" phenotypes from medical imaging beyond traditional clinician interpretation. Using large echocardiography datasets from two healthcare systems, we test whether it is possible to predict age, race, and sex from cardiac ultrasound images using deep learning algorithms and assess the impact of varying confounding variables. Using a total of 433,469 videos from Cedars-Sinai Medical Center and 99,909 videos from Stanford Medical Center, we trained video-based convolutional neural networks to predict age, sex, and race. We found that deep learning models were able to identify age and sex, while unable to reliably predict race. Without considering confounding differences between categories, the AI model predicted sex with an AUC of 0.85 (95% CI 0.84-0.86), age with a mean absolute error of 9.12 years (95% CI 9.00-9.25), and race with AUCs ranging from 0.63 to 0.71. When predicting race, we show that tuning the proportion of confounding variables (age or sex) in the training data significantly impacts model AUC (ranging from 0.53 to 0.85), while sex and age prediction was not particularly impacted by adjusting race proportion in the training dataset AUC of 0.81-0.83 and 0.80-0.84, respectively. This suggests significant proportion of AI's performance on predicting race could come from confounding features being detected. Further work remains to identify the particular imaging features that associate with demographic information and to better understand the risks of demographic identification in medical AI as it pertains to potentially perpetuating bias and disparities. Abstract Deep learning has been shown to accurately assess “hidden” phenotypes from medical imaging beyond traditional clinician interpretation. Using large echocardiography datasets from two healthcare systems, we test whether it is possible to predict age, race, and sex from cardiac ultrasound images using deep learning algorithms and assess the impact of varying confounding variables. Using a total of 433,469 videos from Cedars-Sinai Medical Center and 99,909 videos from Stanford Medical Center, we trained video-based convolutional neural networks to predict age, sex, and race. We found that deep learning models were able to identify age and sex, while unable to reliably predict race. Without considering confounding differences between categories, the AI model predicted sex with an AUC of 0.85 (95% CI 0.84–0.86), age with a mean absolute error of 9.12 years (95% CI 9.00–9.25), and race with AUCs ranging from 0.63 to 0.71. When predicting race, we show that tuning the proportion of confounding variables (age or sex) in the training data significantly impacts model AUC (ranging from 0.53 to 0.85), while sex and age prediction was not particularly impacted by adjusting race proportion in the training dataset AUC of 0.81–0.83 and 0.80–0.84, respectively. This suggests significant proportion of AI’s performance on predicting race could come from confounding features being detected. Further work remains to identify the particular imaging features that associate with demographic information and to better understand the risks of demographic identification in medical AI as it pertains to potentially perpetuating bias and disparities. |
ArticleNumber | 188 |
Author | Yuan, Neal Duffy, Grant Cheng, Susan Clarke, Shoa L. Christensen, Matthew He, Bryan Ouyang, David |
Author_xml | – sequence: 1 givenname: Grant surname: Duffy fullname: Duffy, Grant organization: Department of Cardiology, Smidt Heart Institute, Cedars-Sinai Medical Center – sequence: 2 givenname: Shoa L. orcidid: 0000-0002-6592-1172 surname: Clarke fullname: Clarke, Shoa L. organization: Division of Cardiovascular Medicine, Department of Medicine, Stanford University – sequence: 3 givenname: Matthew surname: Christensen fullname: Christensen, Matthew organization: Department of Cardiology, Smidt Heart Institute, Cedars-Sinai Medical Center – sequence: 4 givenname: Bryan orcidid: 0000-0002-6150-761X surname: He fullname: He, Bryan organization: Department of Computer Science, Stanford University – sequence: 5 givenname: Neal orcidid: 0000-0001-5782-7437 surname: Yuan fullname: Yuan, Neal organization: San Francisco Veteran Affairs Medical Center, University of California San Francisco – sequence: 6 givenname: Susan orcidid: 0000-0002-4977-036X surname: Cheng fullname: Cheng, Susan organization: Department of Cardiology, Smidt Heart Institute, Cedars-Sinai Medical Center – sequence: 7 givenname: David orcidid: 0000-0002-3813-7518 surname: Ouyang fullname: Ouyang, David email: David.Ouyang@cshs.org organization: Department of Cardiology, Smidt Heart Institute, Cedars-Sinai Medical Center, Division of Artificial Intelligence in Medicine, Department of Medicine, Cedars-Sinai Medical Center |
BackLink | https://www.ncbi.nlm.nih.gov/pubmed/36550271$$D View this record in MEDLINE/PubMed |
BookMark | eNp9kU-PFCEQxYlZ467rfgEPphMvXlqhgAYuJpvx3ySbeNEzoYHuZdIDI3Sb-O1lptd114MnKtTvvaJ4z9FZTNEj9JLgtwRT-a4wIljXYoAWYwG4lU_QBVAl245yOHtQn6OrUnYYY8BMKtY9Q-e04xyDIBfowybFIS3R-VyavXfBzL653jaHXGs7hxSbNDTO79OYzeE22NKEeAKtmZqwN2OI4wv0dDBT8Vd35yX6_unjt82X9ubr5-3m-qa1nOG57Z30QiniHRmUpw4k45b2woCkmFmnvGPO8X4AUtEegChCpcHcVpWVQC_RdvV1yez0Idfx-ZdOJujTRcqjNnkOdvIaDIfei076zjFimbG-lx3rHQVmBajq9X71Oix9Xcf6OGczPTJ93InhVo_pp1ZCYsp5NXhzZ5DTj8WXWe9DsX6aTPRpKRoEl1gpio_o63_QXVpyrF91pDopOMiuUrBSNqdSsh_uH0OwPmau18x1zVyfMteyil49XONe8ifhCtAVKLUVR5__zv6P7W810LiH |
CitedBy_id | crossref_primary_10_1001_jamacardio_2023_3701 crossref_primary_10_5604_01_3001_0053_9338 crossref_primary_10_2214_AJR_22_28802 crossref_primary_10_1093_eurheartj_ehae415 crossref_primary_10_1016_j_cell_2023_01_035 crossref_primary_10_1093_ehjci_jeae037 crossref_primary_10_1093_ehjdh_ztae014 crossref_primary_10_1007_s10278_024_01183_x crossref_primary_10_1186_s12933_023_01985_3 crossref_primary_10_1007_s10278_023_00914_w crossref_primary_10_1016_j_jacc_2024_05_003 crossref_primary_10_1016_j_jacadv_2024_100889 crossref_primary_10_1161_CIRCIMAGING_123_015495 crossref_primary_10_1016_j_echo_2024_04_017 crossref_primary_10_1126_science_adh4260 crossref_primary_10_1016_j_ebiom_2024_105075 |
Cites_doi | 10.1186/1471-2342-9-2 10.1007/s11263-015-0816-y 10.1056/NEJMp2116224 10.1016/j.jcmg.2021.01.008 10.1038/s41591-020-01192-7 10.1161/CIRCIMAGING.108.819938 10.1126/science.aax2342 10.1038/s41591-021-01614-0 10.1016/j.echo.2019.08.012 10.1038/s41551-018-0195-0 10.1038/s41746-019-0216-8 10.1016/j.echo.2020.06.008 10.1161/CIRCEP.119.007284 10.1056/NEJMms2025396 10.1038/d41586-018-05707-8 10.1038/s41586-020-2145-8 10.1056/NEJMms2004740 10.1038/s41591-021-01312-x 10.1186/s12968-020-00683-3 10.1109/CVPR.2016.308 10.1038/s41591-022-01987-w 10.1142/9789811232701_0022 10.1016/j.ebiom.2021.103613 10.1007/s10140-021-01953-y 10.1109/CVPR.2016.90 10.1001/jamacardio.2021.6059 |
ContentType | Journal Article |
Copyright | The Author(s) 2022 2022. The Author(s). The Author(s) 2022. This work is published under http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License. |
Copyright_xml | – notice: The Author(s) 2022 – notice: 2022. The Author(s). – notice: The Author(s) 2022. This work is published under http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License. |
DBID | C6C NPM AAYXX CITATION 3V. 7RV 7X7 7XB 8FI 8FJ 8FK ABUWG AFKRA AZQEC BENPR CCPQU DWQXO FYUFA GHDGH K9. KB0 M0S NAPCQ PIMPY PQEST PQQKQ PQUKI PRINS 7X8 5PM DOA |
DOI | 10.1038/s41746-022-00720-8 |
DatabaseName | Springer Nature OA Free Journals PubMed CrossRef ProQuest Central (Corporate) Nursing & Allied Health Database (ProQuest) Health & Medical Collection ProQuest Central (purchase pre-March 2016) Hospital Premium Collection Hospital Premium Collection (Alumni Edition) ProQuest Central (Alumni) (purchase pre-March 2016) ProQuest Central (Alumni) ProQuest Central ProQuest Central Essentials AUTh Library subscriptions: ProQuest Central ProQuest One Community College ProQuest Central Korea Health Research Premium Collection Health Research Premium Collection (Alumni) ProQuest Health & Medical Complete (Alumni) Nursing & Allied Health Database (Alumni Edition) Health & Medical Collection (Alumni Edition) Nursing & Allied Health Premium Publicly Available Content (ProQuest) ProQuest One Academic Eastern Edition (DO NOT USE) ProQuest One Academic ProQuest One Academic UKI Edition ProQuest Central China MEDLINE - Academic PubMed Central (Full Participant titles) Directory of Open Access Journals |
DatabaseTitle | PubMed CrossRef Publicly Available Content Database ProQuest Central Essentials ProQuest One Academic Eastern Edition ProQuest Health & Medical Complete (Alumni) ProQuest Central (Alumni Edition) ProQuest One Community College ProQuest Nursing & Allied Health Source ProQuest Hospital Collection Health Research Premium Collection (Alumni) ProQuest Central China ProQuest Hospital Collection (Alumni) ProQuest Central Nursing & Allied Health Premium ProQuest Health & Medical Complete Health Research Premium Collection ProQuest One Academic UKI Edition Health and Medicine Complete (Alumni Edition) ProQuest Central Korea ProQuest Nursing & Allied Health Source (Alumni) ProQuest One Academic ProQuest Central (Alumni) MEDLINE - Academic |
DatabaseTitleList | PubMed CrossRef Publicly Available Content Database |
Database_xml | – sequence: 1 dbid: C6C name: SpringerOpen website url: http://www.springeropen.com/ sourceTypes: Publisher – sequence: 2 dbid: DOA name: DOAJ Directory of Open Access Journals url: https://www.doaj.org/ sourceTypes: Open Website – sequence: 3 dbid: NPM name: PubMed url: https://proxy.k.utb.cz/login?url=http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed sourceTypes: Index Database – sequence: 4 dbid: 7X7 name: Health & Medical Collection (Proquest) url: https://search.proquest.com/healthcomplete sourceTypes: Aggregation Database |
DeliveryMethod | fulltext_linktorsrc |
Discipline | Medicine |
EISSN | 2398-6352 |
EndPage | 188 |
ExternalDocumentID | oai_doaj_org_article_2a52be768e6d41c4aceb864bd324c729 10_1038_s41746_022_00720_8 36550271 |
Genre | Journal Article |
GroupedDBID | 0R~ 53G 7RV 7X7 8FI 8FJ AAJSJ ABUWG ACGFS ACSMW ADBBV AFKRA AJTQC ALIPV ALMA_UNASSIGNED_HOLDINGS AOIJS BCNDV BENPR C6C CCPQU EBLON EBS EIHBH FYUFA GROUPED_DOAJ HMCUK HYE M~E NAO NAPCQ NO~ OK1 PGMZT PIMPY RNT RPM SNYQT UKHRP NPM AAYXX CITATION 3V. 7XB 8FK AZQEC DWQXO K9. PQEST PQQKQ PQUKI PRINS 7X8 5PM |
ID | FETCH-LOGICAL-c540t-bd8e7991ed1f9e3d2845c3b7a28304cd9ed4dd5bf21bd8b2219138a05c799c823 |
IEDL.DBID | RPM |
ISSN | 2398-6352 |
IngestDate | Tue Oct 22 15:13:31 EDT 2024 Tue Sep 17 21:32:20 EDT 2024 Fri Oct 25 01:28:38 EDT 2024 Thu Oct 10 18:39:02 EDT 2024 Fri Nov 22 00:48:18 EST 2024 Sat Sep 28 08:19:29 EDT 2024 Fri Oct 11 20:44:51 EDT 2024 |
IsDoiOpenAccess | true |
IsOpenAccess | true |
IsPeerReviewed | true |
IsScholarly | true |
Issue | 1 |
Language | English |
License | 2022. The Author(s). Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. |
LinkModel | DirectLink |
MergedId | FETCHMERGED-LOGICAL-c540t-bd8e7991ed1f9e3d2845c3b7a28304cd9ed4dd5bf21bd8b2219138a05c799c823 |
Notes | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 |
ORCID | 0000-0002-3813-7518 0000-0001-5782-7437 0000-0002-6592-1172 0000-0002-6150-761X 0000-0002-4977-036X |
OpenAccessLink | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9780355/ |
PMID | 36550271 |
PQID | 2756875286 |
PQPubID | 5061815 |
PageCount | 1 |
ParticipantIDs | doaj_primary_oai_doaj_org_article_2a52be768e6d41c4aceb864bd324c729 pubmedcentral_primary_oai_pubmedcentral_nih_gov_9780355 proquest_miscellaneous_2758099305 proquest_journals_2756875286 crossref_primary_10_1038_s41746_022_00720_8 pubmed_primary_36550271 springer_journals_10_1038_s41746_022_00720_8 |
PublicationCentury | 2000 |
PublicationDate | 2022-12-22 |
PublicationDateYYYYMMDD | 2022-12-22 |
PublicationDate_xml | – month: 12 year: 2022 text: 2022-12-22 day: 22 |
PublicationDecade | 2020 |
PublicationPlace | London |
PublicationPlace_xml | – name: London – name: England |
PublicationTitle | NPJ digital medicine |
PublicationTitleAbbrev | npj Digit. Med |
PublicationTitleAlternate | NPJ Digit Med |
PublicationYear | 2022 |
Publisher | Nature Publishing Group UK Nature Publishing Group Nature Portfolio |
Publisher_xml | – name: Nature Publishing Group UK – name: Nature Publishing Group – name: Nature Portfolio |
References | Cheng (CR20) 2009; 2 Ghorbani (CR8) 2020; 3 Obermeyer, Powers, Vogeli, Mullainathan (CR17) 2019; 366 CR19 CR14 CR12 Rajpurkar, Chen, Banerjee, Topol (CR1) 2022; 28 Poplin (CR6) 2018; 2 Raghu, Weiss, Hoffmann, Aerts, Lu (CR10) 2021; 14 Kawel-Boehm (CR23) 2020; 22 Zou, Schiebinger (CR13) 2018; 559 Russakovsky (CR2) 2015; 115 Deyrup, Graves (CR25) 2022; 386 CR4 Ouyang (CR18) 2020; 580 CR3 Vyas, Eisenstein, Jones (CR27) 2020; 383 CR5 Attia (CR11) 2019; 12 Cain (CR22) 2009; 9 CR7 CR29 CR9 Wu (CR16) 2021; 27 CR26 Asch (CR21) 2019; 32 Bailey, Feldman, Bassett (CR28) 2021; 384 Pierson, Cutler, Leskovec, Mullainathan, Obermeyer (CR15) 2021; 27 Miyoshi (CR24) 2020; 33 P Rajpurkar (720_CR1) 2022; 28 A Ghorbani (720_CR8) 2020; 3 720_CR12 720_CR4 720_CR3 720_CR5 720_CR7 720_CR14 E Wu (720_CR16) 2021; 27 J Zou (720_CR13) 2018; 559 E Pierson (720_CR15) 2021; 27 A Deyrup (720_CR25) 2022; 386 D Ouyang (720_CR18) 2020; 580 ZD Bailey (720_CR28) 2021; 384 DA Vyas (720_CR27) 2020; 383 720_CR29 R Poplin (720_CR6) 2018; 2 T Miyoshi (720_CR24) 2020; 33 FM Asch (720_CR21) 2019; 32 PA Cain (720_CR22) 2009; 9 720_CR26 ZI Attia (720_CR11) 2019; 12 S Cheng (720_CR20) 2009; 2 Z Obermeyer (720_CR17) 2019; 366 VK Raghu (720_CR10) 2021; 14 720_CR9 720_CR19 O Russakovsky (720_CR2) 2015; 115 N Kawel-Boehm (720_CR23) 2020; 22 |
References_xml | – volume: 9 year: 2009 ident: CR22 article-title: Age and gender specific normal values of left ventricular mass, volume and function for gradient echo magnetic resonance imaging: a cross sectional study publication-title: BMC Med. Imaging doi: 10.1186/1471-2342-9-2 contributor: fullname: Cain – volume: 115 start-page: 211 year: 2015 end-page: 252 ident: CR2 article-title: ImageNet large scale visual recognition challenge publication-title: Int. J. Comput. Vis. doi: 10.1007/s11263-015-0816-y contributor: fullname: Russakovsky – volume: 386 start-page: 501 year: 2022 end-page: 503 ident: CR25 article-title: Racial biology and medical misconceptions publication-title: N. Engl. J. Med. doi: 10.1056/NEJMp2116224 contributor: fullname: Graves – ident: CR4 – volume: 14 start-page: 2226 year: 2021 end-page: 2236 ident: CR10 article-title: Deep learning to estimate biological age from chest radiographs publication-title: JACC Cardiovasc. Imaging doi: 10.1016/j.jcmg.2021.01.008 contributor: fullname: Lu – ident: CR14 – volume: 27 start-page: 136 year: 2021 end-page: 140 ident: CR15 article-title: An algorithmic approach to reducing unexplained pain disparities in underserved populations publication-title: Nat. Med. doi: 10.1038/s41591-020-01192-7 contributor: fullname: Obermeyer – ident: CR12 – volume: 2 start-page: 191 year: 2009 end-page: 198 ident: CR20 article-title: Age-related left ventricular remodeling and associated risk for cardiovascular outcomes: the Multi-Ethnic Study of Atherosclerosis publication-title: Circ. Cardiovasc. Imaging doi: 10.1161/CIRCIMAGING.108.819938 contributor: fullname: Cheng – volume: 366 start-page: 447 year: 2019 end-page: 453 ident: CR17 article-title: Dissecting racial bias in an algorithm used to manage the health of populations publication-title: Science doi: 10.1126/science.aax2342 contributor: fullname: Mullainathan – volume: 28 start-page: 31 year: 2022 end-page: 38 ident: CR1 article-title: AI in health and medicine publication-title: Nat. Med. doi: 10.1038/s41591-021-01614-0 contributor: fullname: Topol – ident: CR29 – volume: 32 start-page: 1396.e2 year: 2019 end-page: 1406.e2 ident: CR21 article-title: Similarities and differences in left ventricular size and function among races and nationalities: results of the World Alliance Societies of Echocardiography Normal Values Study publication-title: J. Am. Soc. Echocardiogr. doi: 10.1016/j.echo.2019.08.012 contributor: fullname: Asch – volume: 2 start-page: 158 year: 2018 end-page: 164 ident: CR6 article-title: Prediction of cardiovascular risk factors from retinal fundus photographs via deep learning publication-title: Nat. Biomed. Eng. doi: 10.1038/s41551-018-0195-0 contributor: fullname: Poplin – volume: 3 start-page: 10 year: 2020 ident: CR8 article-title: Deep learning interpretation of echocardiograms publication-title: NPJ Digit Med doi: 10.1038/s41746-019-0216-8 contributor: fullname: Ghorbani – ident: CR19 – ident: CR3 – volume: 33 start-page: 1223 year: 2020 end-page: 1233 ident: CR24 article-title: Left ventricular diastolic function in healthy adult individuals: results of the World Alliance Societies of Echocardiography Normal Values Study publication-title: J. Am. Soc. Echocardiogr. doi: 10.1016/j.echo.2020.06.008 contributor: fullname: Miyoshi – volume: 12 start-page: e007284 year: 2019 ident: CR11 article-title: Age and sex estimation using artificial intelligence from standard 12-lead ECGs publication-title: Circ. Arrhythm. Electrophysiol. doi: 10.1161/CIRCEP.119.007284 contributor: fullname: Attia – volume: 384 start-page: 768 year: 2021 end-page: 773 ident: CR28 article-title: How structural racism works - racist policies as a root cause of U.s. racial health inequities publication-title: N. Engl. J. Med. doi: 10.1056/NEJMms2025396 contributor: fullname: Bassett – ident: CR9 – volume: 559 start-page: 324 year: 2018 end-page: 326 ident: CR13 article-title: AI can be sexist and racist — it’s time to make it fair publication-title: Nature doi: 10.1038/d41586-018-05707-8 contributor: fullname: Schiebinger – volume: 580 start-page: 252 year: 2020 end-page: 256 ident: CR18 article-title: Video-based AI for beat-to-beat assessment of cardiac function publication-title: Nature doi: 10.1038/s41586-020-2145-8 contributor: fullname: Ouyang – ident: CR5 – ident: CR7 – volume: 383 start-page: 874 year: 2020 end-page: 882 ident: CR27 article-title: Hidden in plain sight - reconsidering the use of race correction in clinical algorithms publication-title: N. Engl. J. Med. doi: 10.1056/NEJMms2004740 contributor: fullname: Jones – volume: 27 start-page: 582 year: 2021 end-page: 584 ident: CR16 article-title: How medical AI devices are evaluated: limitations and recommendations from an analysis of FDA approvals publication-title: Nat. Med. doi: 10.1038/s41591-021-01312-x contributor: fullname: Wu – ident: CR26 – volume: 22 start-page: 1 year: 2020 end-page: 63 ident: CR23 article-title: Reference ranges (“normal values”) for cardiovascular magnetic resonance (CMR) in adults and children: 2020 update publication-title: J. Cardiovasc. Magn. Reson. doi: 10.1186/s12968-020-00683-3 contributor: fullname: Kawel-Boehm – ident: 720_CR3 doi: 10.1109/CVPR.2016.308 – volume: 2 start-page: 191 year: 2009 ident: 720_CR20 publication-title: Circ. Cardiovasc. Imaging doi: 10.1161/CIRCIMAGING.108.819938 contributor: fullname: S Cheng – ident: 720_CR26 doi: 10.1038/s41591-022-01987-w – volume: 559 start-page: 324 year: 2018 ident: 720_CR13 publication-title: Nature doi: 10.1038/d41586-018-05707-8 contributor: fullname: J Zou – volume: 366 start-page: 447 year: 2019 ident: 720_CR17 publication-title: Science doi: 10.1126/science.aax2342 contributor: fullname: Z Obermeyer – volume: 14 start-page: 2226 year: 2021 ident: 720_CR10 publication-title: JACC Cardiovasc. Imaging doi: 10.1016/j.jcmg.2021.01.008 contributor: fullname: VK Raghu – volume: 27 start-page: 136 year: 2021 ident: 720_CR15 publication-title: Nat. Med. doi: 10.1038/s41591-020-01192-7 contributor: fullname: E Pierson – volume: 115 start-page: 211 year: 2015 ident: 720_CR2 publication-title: Int. J. Comput. Vis. doi: 10.1007/s11263-015-0816-y contributor: fullname: O Russakovsky – volume: 28 start-page: 31 year: 2022 ident: 720_CR1 publication-title: Nat. Med. doi: 10.1038/s41591-021-01614-0 contributor: fullname: P Rajpurkar – volume: 384 start-page: 768 year: 2021 ident: 720_CR28 publication-title: N. Engl. J. Med. doi: 10.1056/NEJMms2025396 contributor: fullname: ZD Bailey – volume: 386 start-page: 501 year: 2022 ident: 720_CR25 publication-title: N. Engl. J. Med. doi: 10.1056/NEJMp2116224 contributor: fullname: A Deyrup – volume: 27 start-page: 582 year: 2021 ident: 720_CR16 publication-title: Nat. Med. doi: 10.1038/s41591-021-01312-x contributor: fullname: E Wu – ident: 720_CR29 – volume: 3 start-page: 10 year: 2020 ident: 720_CR8 publication-title: NPJ Digit Med doi: 10.1038/s41746-019-0216-8 contributor: fullname: A Ghorbani – ident: 720_CR14 doi: 10.1142/9789811232701_0022 – volume: 2 start-page: 158 year: 2018 ident: 720_CR6 publication-title: Nat. Biomed. Eng. doi: 10.1038/s41551-018-0195-0 contributor: fullname: R Poplin – ident: 720_CR7 doi: 10.1016/j.ebiom.2021.103613 – ident: 720_CR9 doi: 10.1007/s10140-021-01953-y – volume: 32 start-page: 1396.e2 year: 2019 ident: 720_CR21 publication-title: J. Am. Soc. Echocardiogr. doi: 10.1016/j.echo.2019.08.012 contributor: fullname: FM Asch – volume: 383 start-page: 874 year: 2020 ident: 720_CR27 publication-title: N. Engl. J. Med. doi: 10.1056/NEJMms2004740 contributor: fullname: DA Vyas – volume: 9 year: 2009 ident: 720_CR22 publication-title: BMC Med. Imaging doi: 10.1186/1471-2342-9-2 contributor: fullname: PA Cain – volume: 580 start-page: 252 year: 2020 ident: 720_CR18 publication-title: Nature doi: 10.1038/s41586-020-2145-8 contributor: fullname: D Ouyang – volume: 33 start-page: 1223 year: 2020 ident: 720_CR24 publication-title: J. Am. Soc. Echocardiogr. doi: 10.1016/j.echo.2020.06.008 contributor: fullname: T Miyoshi – ident: 720_CR12 – ident: 720_CR4 doi: 10.1109/CVPR.2016.90 – ident: 720_CR19 doi: 10.1001/jamacardio.2021.6059 – ident: 720_CR5 – volume: 12 start-page: e007284 year: 2019 ident: 720_CR11 publication-title: Circ. Arrhythm. Electrophysiol. doi: 10.1161/CIRCEP.119.007284 contributor: fullname: ZI Attia – volume: 22 start-page: 1 year: 2020 ident: 720_CR23 publication-title: J. Cardiovasc. Magn. Reson. doi: 10.1186/s12968-020-00683-3 contributor: fullname: N Kawel-Boehm |
SSID | ssj0002048946 |
Score | 2.3876815 |
Snippet | Deep learning has been shown to accurately assess “hidden” phenotypes from medical imaging beyond traditional clinician interpretation. Using large... Deep learning has been shown to accurately assess "hidden" phenotypes from medical imaging beyond traditional clinician interpretation. Using large... Abstract Deep learning has been shown to accurately assess “hidden” phenotypes from medical imaging beyond traditional clinician interpretation. Using large... |
SourceID | doaj pubmedcentral proquest crossref pubmed springer |
SourceType | Open Website Open Access Repository Aggregation Database Index Database Publisher |
StartPage | 188 |
SubjectTerms | 631/114/1305 631/114/2397 Age Artificial intelligence Biomedicine Biotechnology Confounding (Statistics) Deep learning Demographics Medical imaging Medicine Medicine & Public Health |
SummonAdditionalLinks | – databaseName: Directory of Open Access Journals dbid: DOA link: http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV1LT9wwEB4hDlUvFS2lDYXKSNxai_iRrHPkUQRIcAKJm-XHRCwSWcQu_5-xk11YWtQL13gijebz4xuN_Q3ArlZOC68Cl8JLrkUbuDfB8VCijKXDWpXpofD5RX1ypc-uq-sXrb7SnbBeHrgP3J50lfRIpBjrqEXQLqA3tfaRmEAgZph331K-SKZuc3lNm0bXwyuZUpm9qSbune7bSp7Usktulk6iLNj_L5b592XJVxXTfBAdr8GngUGy_d7zz7CC3Rf4cD7UyNfhKD3iS82SiNix_DBkhmz_lN0_JJOEA5u0LOJdL1Y9DlM27rIhwcXGd7lt0Ve4Ov5zeXjCh14JPBDnmnEfDY6I62EUbYMq0qlTBeVHLgl86RAbjDrGyreESTRe0kYllHFlFeivYKTagNVu0uF3YCEIJaJrnGobyr584yltUa1ApxosVSzg1zxu9r6XxLC5lK2M7aNsKco2R9maAg5SaBeWSc46fyCQ7QCy_R_IBWzNgbHDGpvaJFxP2ZY0dQE7i2FaHank4TqcPGYbQxyYNrUCvvU4LjxRNWVnciQKGC0hvOTq8kg3vskK3Em2iYhaAb_nc-HZrbdDsfkeofgBH2WaxEJyKbdgdfbwiNvEi2b-Z14CT42mCUA priority: 102 providerName: Directory of Open Access Journals – databaseName: Health & Medical Collection dbid: 7X7 link: http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwfV1Lb9QwEB5BkRAX1PJMH8hI3MBq_EjinKryqApSOVFpb5ZfgT00WXa3_58ZJ7vV8romE8mZGXu-8djfALzRymnhVeBSeMm16AL3JjgeyiRj6VKtSroofPW1vrzWX2bVbNpwW03HKjdrYl6o4xBoj_yUaMoRW0tTny1-cuoaRdXVqYXGfXggJIZy9Odm1mz3WIiUttX1dFemVOZ0pRGB06lbyYkzu-RmJx5l2v6_Yc0_j0z-VjfN4ehiHx5POJKdj4Y_gHupfwIPr6ZK-VP4SFf5qGUSwjuWr4esEzv_zBZLEiFrsKFjMd2MlNXzsGLzPgui0dj8JjcvegbXF5--fbjkU8cEHhB5rbmPJjWI-FIUXZtUxNhTBeUbRzRfOsQ2RR1j5Tu0TDRe4nIllHFlFfCrYKR6Dnv90KeXwEIQSkTXOtW1mIP51mPyojqRnGpTqWIBbzd6s4uRGMPmgrYydtSyRS3brGVrCnhPqt1KEql1fjAsv9tpjljpKukT5j-pjloE7ULyptY-IugLmAQUcLwxjJ1m2sre-UUBr7evcY5Q4cP1abjNMgaRMC5tBbwY7bgdiaoxR5ONKKDZsfDOUHff9PMfmYebyJsQrhXwbuMLd8P6tyoO__8XR_BIknsKyaU8hr318jadIO5Z-1fZuX8B73UAZA priority: 102 providerName: ProQuest – databaseName: Springer Nature OA Free Journals dbid: C6C link: http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LT9wwEB5RkKpeUEtbSEsrI_XWWo0fSZwjbIsoEpyKxM3yK-oeyCJ2-f_MONmt0sKBazyWrJmx55uM5zPAF62cFl4FLoWXXIsucG-C46FMMpYu1aqkRuGLy_rsSp9fV9cjTQ71wkzq98p8X2qEzHRNVnIiuS65eQE7GIMb8uZZPdv8TyEC2lbXY1_M41MnsSdT9D-GK_-_HvlPjTSHntPXsDtiRnY8GPkNbKV-D15ejFXxt_CD2vboeSSEciy3gqwSO_7Fbu9IhDTPFh2L6Wagp56HJZv3WRANxOY3-aGid3B1-vP37IyPryPwgChrxX00qUF0l6Lo2qQixpkqKN84ovTSIbYp6hgr36EVovESjyahjCurgLOCkeo9bPeLPh0AC0EoEV3rVNdivuVbj4mK6kRyqk2ligV8XevN3g4kGDYXr5Wxg5YtatlmLVtTwAmpdiNJBNb5A9rVjvvBSldJnzDXSXXUImgXkje19hEBXkDAX8Dh2jB23FVLS1T1mF9JUxdwtBnG_UBFDtenxX2WMYh68RgrYH-w42YlqsZ8TDaigGZi4clSpyP9_E_m3CaiJoRmBXxb-8LfZT2tig_PE_8IryS5q5BcykPYXt3dp0-IeVb-c3b2B6DN-RM priority: 102 providerName: Springer Nature |
Title | Confounders mediate AI prediction of demographics in medical imaging |
URI | https://link.springer.com/article/10.1038/s41746-022-00720-8 https://www.ncbi.nlm.nih.gov/pubmed/36550271 https://www.proquest.com/docview/2756875286 https://search.proquest.com/docview/2758099305 https://pubmed.ncbi.nlm.nih.gov/PMC9780355 https://doaj.org/article/2a52be768e6d41c4aceb864bd324c729 |
Volume | 5 |
hasFullText | 1 |
inHoldings | 1 |
isFullTextHit | |
isPrint | |
link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV1Lb9NAEB61RUJcEG9cSmQkbrCN92F7fUxDqxIpVQVUym21L0MQcaIk_f-dXduB8LhwsrRey6uZWc83nplvAd4KrgU13BJGDSOC1pYYaTWxmWcu077gWWgUnl4VlzdiMstnB5D3vTCxaN-a-WnzY3HazL_F2srVwg77OrHh9XQcWHPQTw4P4RDd7y8h-veYWROyEkXXIJNxOdwIhN2h1JaRQJSdkXBIHy8Qm7OS7vmjSNv_N6z5Z8nkb3nT6I4uHsHDDkemo3a9j-HAN0_g_rTLlD-FD6GVLxyZhPAuje0hW5-OPqardZgStJEu69T5RUtZPbebdN7Eiai0dL6Ihxc9g5uL8y_jS9KdmEAsIq8tMU76EhGfd7SuPHfoe3LLTakDzZewrvJOOJebGjXjpGH4uaJc6iy3-JSVjD-Ho2bZ-JeQWks5dbrSvK4wBjOVweCF19RrXvmMuwTe9XJTq5YYQ8WENpeqFbhCgasocCUTOAui3c0MpNZxYLn-qjrVKqZzZjzGP75wglqhrTeyEMYh6LMYBCRw0itGdTttowJ9PcZcTBYJvNndxj0SEh-68cvbOEciEsZPWwIvWj3uVtLbQQLlnob3lrp_B80y8nB3ZpjA-94Wfi7r36I4_u8XvYIHLBgxZYSxEzjarm_9a4REWzPAjTArB3BvNJp8nuD17Pzq-hOOjovxIP5mGMRNcgebew8s |
link.rule.ids | 230,314,727,780,784,864,885,2102,12056,21388,27924,27925,31719,31720,33744,33745,41120,42189,43310,43805,51576,53791,53793 |
linkProvider | National Library of Medicine |
linkToHtml | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwfV1Lb9swDCa2FNh2KfbsvLabB-y2CbUeduRT0a4t0q0JhqEFehP08pZD7SxJ_38pWUmRva42DcgkRX0UpY8AHwTXghpuCaOGEUEbS4y0mtjCM1doX_EiXBQeT6rRlfhyXV6nDbdFOla5iokxULvOhj3yg0BTjtiayepw9ouErlGhuppaaDyErcCcXg5g6_h08u37epcl0NLWokq3ZQouDxYCMXg4d8tIYM0uiNxYkSJx_9_Q5p-HJn-rnMYF6ewpbCckmR_1pn8GD3z7HB6NU638BZyEy3yhaRICvDxeEFn6_Og8n82DSLBH3jW58zc9afXULvJpGwXRbPn0JrYveglXZ6eXn0ck9UwgFrHXkhgn_RAxn3e0qT13uPqUlpuhDkRfwrraO-FcaRq0jZOGYcCiXOqitPiVlYy_gkHbtf415NZSTp2uNW9qzMJMbTB94Q31mte-4C6Djyu9qVlPjaFiSZtL1WtZoZZV1LKSGRwH1a4lA611fNDNf6g0SxTTJTMeMyBfOUGt0NYbWQnjEPZZTAMy2FsZRqW5tlD3npHB-_VrnCWh9KFb391GGYlYGINbBju9Hdcj4RVmaWxIMxhuWHhjqJtv2unPyMQd6JsQsGXwaeUL98P6tyre_P8v3sHj0eX4Ql2cT77uwhMWXJUywtgeDJbzW7-PKGhp3iZXvwOF1wS6 |
linkToPdf | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1Lb9QwEB5BkSouiDeBAkbiBlbjR7LOsWxZtUArDlTqzfJjAntodrWb_n_GTnZhoRy4xmPJmhnb32Q83wC81cpp4VXgUnjJtWgD9yY4HkqUsXRYqzIVCp-d1ycX-tNldflbFX9-7b5JSQ41DYmlqesPl7Edi8TN4VoTkE6PZyVP1NclN7fhjibokJo3TOvp9i9LoqVtdD1Wy9w8dedGysT9N6HNvx9N_pE5zRfS7D7cG5EkOxpM_wBuYfcQ9s_GXPkjOE7FfKlpEgE8lgtEemRHp2y5SiLJHmzRsohXA2n1PKzZvMuCZDY2v8rtix7Dxezjt-kJH3sm8EDYq-c-GpwQ5sMo2gZVpNunCspPXCL60iE2GHWMlW_JNtF4SQeWUMaVVaBZwUj1BPa6RYfPgIUglIiucaptKArzjafwRbUCnWqwVLGAdxu92eVAjWFzSlsZO2jZkpZt1rI1BXxIqt1KJlrr_GGx-m7HXWKlq6RHioCwjloE7QJ6U2sfCfYFCgMKONgYxo57bW0TgT1FXdLUBbzZDtMuSakP1-HiOssYwsJ0uBXwdLDjdiWqpihNTkQBkx0L7yx1d6Sb_8hM3Im-iQBbAe83vvBrWf9WxfP_E38N-1-PZ_bL6fnnF3BXJs8Vkkt5AHv96hpfEijq_avs9z8BXAAESg |
openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Confounders+mediate+AI+prediction+of+demographics+in+medical+imaging&rft.jtitle=NPJ+digital+medicine&rft.au=Duffy%2C+Grant&rft.au=Clarke%2C+Shoa+L&rft.au=Christensen%2C+Matthew&rft.au=He%2C+Bryan&rft.date=2022-12-22&rft.eissn=2398-6352&rft.volume=5&rft.issue=1&rft.spage=188&rft_id=info:doi/10.1038%2Fs41746-022-00720-8&rft_id=info%3Apmid%2F36550271&rft.externalDocID=36550271 |
thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2398-6352&client=summon |
thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2398-6352&client=summon |
thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2398-6352&client=summon |