دورية أكاديمية
Deep neural networks reveal novel sex-specific electrocardiographic features relevant for mortality risk.
| العنوان: | Deep neural networks reveal novel sex-specific electrocardiographic features relevant for mortality risk. |
|---|---|
| المؤلفون: | Siegersma KR; Department of Cardiology, Amsterdam University Medical Centres, VU University Amsterdam, Amsterdam, The Netherlands.; Laboratory of Experimental Cardiology, University Medical Center Utrecht, Utrecht University, Utrecht, The Netherlands., van de Leur RR; Department of Cardiology, University Medical Center Utrecht, Utrecht, The Netherlands.; Netherlands Heart Institute, Utrecht, The Netherlands., Onland-Moret NC; Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht University, Utrecht, The Netherlands., Leon DA; Faculty of Epidemiology and Population Health, London School of Hygiene & Tropical Medicine, London WC1E 7HT, UK.; International Laboratory for Population and Health, National Research University, Higher School of Economics, Moscow 101000, Russian Federation.; Department of Community Medicine, UiT The Arctic University of Norway, Tromsø, Norway., Diez-Benavente E; Laboratory of Experimental Cardiology, University Medical Center Utrecht, Utrecht University, Utrecht, The Netherlands., Rozendaal L; Julius Gezondheidscentrum Parkwijk, Utrecht, The Netherlands., Bots ML; Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht University, Utrecht, The Netherlands., Coronel R; Heart Center, Department of Experimental Cardiology, AMC, Amsterdam University Medical Centres, Amsterdam, The Netherlands., Appelman Y; Department of Cardiology, Amsterdam University Medical Centres, VU University Amsterdam, Amsterdam, The Netherlands., Hofstra L; Department of Cardiology, Amsterdam University Medical Centres, VU University Amsterdam, Amsterdam, The Netherlands.; Cardiology Centers of the Netherlands, Amsterdam, The Netherlands., van der Harst P; Department of Cardiology, University Medical Center Utrecht, Utrecht, The Netherlands., Doevendans PA; Department of Cardiology, University Medical Center Utrecht, Utrecht, The Netherlands.; Netherlands Heart Institute, Utrecht, The Netherlands., Hassink RJ; Department of Cardiology, University Medical Center Utrecht, Utrecht, The Netherlands., den Ruijter HM; Laboratory of Experimental Cardiology, University Medical Center Utrecht, Utrecht University, Utrecht, The Netherlands., van Es R; Department of Cardiology, University Medical Center Utrecht, Utrecht, The Netherlands. |
| المصدر: | European heart journal. Digital health [Eur Heart J Digit Health] 2022 Mar 21; Vol. 3 (2), pp. 245-254. Date of Electronic Publication: 2022 Mar 21 (Print Publication: 2022). |
| نوع المنشور: | Journal Article |
| اللغة: | English |
| بيانات الدورية: | Publisher: Oxford University Press Country of Publication: England NLM ID: 101778323 Publication Model: eCollection Cited Medium: Internet ISSN: 2634-3916 (Electronic) Linking ISSN: 26343916 NLM ISO Abbreviation: Eur Heart J Digit Health Subsets: PubMed not MEDLINE |
| أسماء مطبوعة: | Original Publication: [Oxford] : Oxford University Press, [2020]- |
| مستخلص: | Aims: Incorporation of sex in study design can lead to discoveries in medical research. Deep neural networks (DNNs) accurately predict sex based on the electrocardiogram (ECG) and we hypothesized that misclassification of sex is an important predictor for mortality. Therefore, we first developed and validated a DNN that classified sex based on the ECG and investigated the outcome. Second, we studied ECG drivers of DNN-classified sex and mortality. Methods and Results: A DNN was trained to classify sex based on 131 673 normal ECGs. The algorithm was validated on internal (68 500 ECGs) and external data sets (3303 and 4457 ECGs). The survival of sex (mis)classified groups was investigated using time-to-event analysis and sex-stratified mediation analysis of ECG features. The DNN successfully distinguished female from male ECGs {internal validation: area under the curve (AUC) 0.96 [95% confidence interval (CI): 0.96, 0.97]; external validations: AUC 0.89 (95% CI: 0.88, 0.90), 0.94 (95% CI: 0.93, 0.94)}. Sex-misclassified individuals (11%) had a 1.4 times higher mortality risk compared with correctly classified peers. The ventricular rate was the strongest mediating ECG variable (41%, 95% CI: 31%, 56%) in males, while the maximum amplitude of the ST segment was strongest in females (18%, 95% CI: 11%, 39%). Short QRS duration was associated with higher mortality risk. Conclusion: Deep neural networks accurately classify sex based on ECGs. While the proportion of ECG-based sex misclassifications is low, it is an interesting biomarker. Investigation of the causal pathway between misclassification and mortality uncovered new ECG features that might be associated with mortality. Increased emphasis on sex as a biological variable in artificial intelligence is warranted. (© The Author(s) 2022. Published by Oxford University Press on behalf of European Society of Cardiology.) |
| References: | J Am Heart Assoc. 2020 May 18;9(10):e015138. (PMID: 32406296) Nat Med. 2019 Jan;25(1):70-74. (PMID: 30617318) J Pers Soc Psychol. 1986 Dec;51(6):1173-82. (PMID: 3806354) Nat Rev Cardiol. 2021 Jul;18(7):465-478. (PMID: 33526938) Lancet. 2021 Jun 19;397(10292):2385-2438. (PMID: 34010613) Circulation. 2019 Mar 19;139(12):1469-1471. (PMID: 30883224) J Am Coll Cardiol. 2020 Feb 25;75(7):722-733. (PMID: 32081280) Nature. 2019 Nov;575(7781):137-146. (PMID: 31695204) Epidemiology. 2011 Sep;22(5):660-70. (PMID: 21709561) Circ Arrhythm Electrophysiol. 2021 Feb;14(2):e009056. (PMID: 33401921) J Electrocardiol. 2014 Nov-Dec;47(6):914-21. (PMID: 25194872) J Hypertens. 2010 Jul;28(7):1527-33. (PMID: 20574251) Arrhythm Electrophysiol Rev. 2020 Nov;9(3):146-154. (PMID: 33240510) Am Heart J. 1987 Jun;113(6):1489-94. (PMID: 3591616) Am J Public Health. 2001 Aug;91(8):1258-63. (PMID: 11499115) Am J Cardiol. 2013 Sep 1;112(5):671-7. (PMID: 23726176) Clin Cardiol. 2017 Oct;40(10):865-872. (PMID: 28605034) Front Physiol. 2021 Mar 23;12:648396. (PMID: 33833689) Wellcome Open Res. 2018 Jun 4;3:67. (PMID: 30123849) Eur J Epidemiol. 2005;20(3):285-7. (PMID: 15921047) Heart Rhythm. 2009 Aug;6(8):1202-8, 1208.e1. (PMID: 19632634) Circulation. 2015 Jan 13;131(2):211-9. (PMID: 25561516) Lancet. 2019 Sep 7;394(10201):861-867. (PMID: 31378392) Ann Noninvasive Electrocardiol. 2012 Oct;17(4):340-8. (PMID: 23094880) J Am Coll Cardiol. 2018 May 8;71(18):1970-1972. (PMID: 29724349) Am J Cardiol. 2017 Apr 1;119(7):1003-1007. (PMID: 28132682) J Am Coll Cardiol. 2018 May 8;71(18):1960-1969. (PMID: 29724348) Int J Cardiol. 2017 Apr 1;232:216-221. (PMID: 28111054) PLoS One. 2020 May 29;15(5):e0233898. (PMID: 32470091) J Intern Med. 2000 Feb;247(2):231-9. (PMID: 10692086) Circ Arrhythm Electrophysiol. 2019 Sep;12(9):e007284. (PMID: 31450977) |
| فهرسة مساهمة: | Keywords: Artificial intelligence; Electrocardiography; Neural network; Sex differences |
| تواريخ الأحداث: | Date Created: 20230130 Latest Revision: 20230202 |
| رمز التحديث: | 20231215 |
| مُعرف محوري في PubMed: | PMC9707888 |
| DOI: | 10.1093/ehjdh/ztac010 |
| PMID: | 36713005 |
| قاعدة البيانات: | MEDLINE |
Full Text Finder | تدمد: | 2634-3916 |
|---|---|
| DOI: | 10.1093/ehjdh/ztac010 |