{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,10]],"date-time":"2026-04-10T16:20:04Z","timestamp":1775838004509,"version":"3.50.1"},"reference-count":37,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2024,2,3]],"date-time":"2024-02-03T00:00:00Z","timestamp":1706918400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2024,2,3]],"date-time":"2024-02-03T00:00:00Z","timestamp":1706918400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["npj Digit. Med."],"abstract":"<jats:title>Abstract<\/jats:title><jats:p>Epicardial adipose tissue (EAT) volume and attenuation are associated with cardiovascular risk, but manual annotation is time-consuming. We evaluated whether automated deep learning-based EAT measurements from ungated computed tomography (CT) are associated with death or myocardial infarction (MI). We included 8781 patients from 4 sites without known coronary artery disease who underwent hybrid myocardial perfusion imaging. Of those, 500 patients from one site were used for model training and validation, with the remaining patients held out for testing (<jats:italic>n<\/jats:italic>\u2009=\u20093511 internal testing, <jats:italic>n<\/jats:italic>\u2009=\u20094770 external testing). We modified an existing deep learning model to first identify the cardiac silhouette, then automatically segment EAT based on attenuation thresholds. Deep learning EAT measurements were obtained in &lt;2\u2009s compared to 15\u2009min for expert annotations. There was excellent agreement between EAT attenuation (Spearman correlation 0.90 internal, 0.82 external) and volume (Spearman correlation 0.90 internal, 0.91 external) by deep learning and expert segmentation in all 3 sites (Spearman correlation 0.90\u20130.98). During median follow-up of 2.7 years (IQR 1.6\u20134.9), 565 patients experienced death or MI. Elevated EAT volume and attenuation were independently associated with an increased risk of death or MI after adjustment for relevant confounders. Deep learning can automatically measure EAT volume and attenuation from low-dose, ungated CT with excellent correlation with expert annotations, but in a fraction of the time. EAT measurements offer additional prognostic insights within the context of hybrid perfusion imaging.<\/jats:p>","DOI":"10.1038\/s41746-024-01020-z","type":"journal-article","created":{"date-parts":[[2024,2,3]],"date-time":"2024-02-03T09:02:14Z","timestamp":1706950934000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":34,"title":["AI-derived epicardial fat measurements improve cardiovascular risk prediction from myocardial perfusion imaging"],"prefix":"10.1038","volume":"7","author":[{"given":"Robert J. H.","family":"Miller","sequence":"first","affiliation":[]},{"given":"Aakash","family":"Shanbhag","sequence":"additional","affiliation":[]},{"given":"Aditya","family":"Killekar","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1531-880X","authenticated-orcid":false,"given":"Mark","family":"Lemley","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3231-1455","authenticated-orcid":false,"given":"Bryan","family":"Bednarski","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4947-9158","authenticated-orcid":false,"given":"Serge D.","family":"Van Kriekinge","sequence":"additional","affiliation":[]},{"given":"Paul B.","family":"Kavanagh","sequence":"additional","affiliation":[]},{"given":"Attila","family":"Feher","sequence":"additional","affiliation":[]},{"given":"Edward J.","family":"Miller","sequence":"additional","affiliation":[]},{"given":"Andrew J.","family":"Einstein","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0686-5449","authenticated-orcid":false,"given":"Terrence D.","family":"Ruddy","sequence":"additional","affiliation":[]},{"given":"Joanna X.","family":"Liang","sequence":"additional","affiliation":[]},{"given":"Valerie","family":"Builoff","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3793-9578","authenticated-orcid":false,"given":"Daniel S.","family":"Berman","sequence":"additional","affiliation":[]},{"given":"Damini","family":"Dey","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6110-938X","authenticated-orcid":false,"given":"Piotr J.","family":"Slomka","sequence":"additional","affiliation":[]}],"member":"297","published-online":{"date-parts":[[2024,2,3]]},"reference":[{"key":"1020_CR1","doi-asserted-by":"publisher","first-page":"2360","DOI":"10.1016\/j.jacc.2018.03.509","volume":"71","author":"M Packer","year":"2018","unstructured":"Packer, M. Epicardial adipose tissue may mediate deleterious effects of obesity and inflammation on the myocardium. J. Am. Coll. Cardiol. 71, 2360\u20132372 (2018).","journal-title":"J. Am. Coll. Cardiol."},{"key":"1020_CR2","doi-asserted-by":"publisher","first-page":"62","DOI":"10.1093\/cvr\/cvv208","volume":"108","author":"B Gaborit","year":"2015","unstructured":"Gaborit, B. et al. Human epicardial adipose tissue has a specific transcriptomic signature depending on its anatomical peri-atrial, peri-ventricular, or peri-coronary location. Cardiovasc. Res. 108, 62\u201373 (2015).","journal-title":"Cardiovasc. Res."},{"key":"1020_CR3","doi-asserted-by":"publisher","first-page":"593","DOI":"10.1038\/s41569-022-00679-9","volume":"19","author":"G Iacobellis","year":"2022","unstructured":"Iacobellis, G. Epicardial adipose tissue in contemporary cardiology. Nat. Rev. Cardiol. 19, 593\u2013606 (2022).","journal-title":"Nat. Rev. Cardiol."},{"key":"1020_CR4","doi-asserted-by":"publisher","first-page":"141","DOI":"10.1016\/j.jcct.2016.01.009","volume":"10","author":"MM Hell","year":"2016","unstructured":"Hell, M. M. et al. Epicardial adipose tissue volume but not density is an independent predictor for myocardial ischemia. J. Cardiovasc. Comput. Tomogr. 10, 141\u2013149 (2016).","journal-title":"J. Cardiovasc. Comput. Tomogr."},{"key":"1020_CR5","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-020-63135-z","volume":"10","author":"N Nerlekar","year":"2020","unstructured":"Nerlekar, N. et al. The natural history of epicardial adipose tissue volume and attenuation: a long-term prospective cohort follow-up study. Sci. Rep. 10, 7109 (2020).","journal-title":"Sci. Rep."},{"key":"1020_CR6","doi-asserted-by":"publisher","first-page":"67","DOI":"10.1016\/j.jcct.2017.11.007","volume":"12","author":"M Goeller","year":"2018","unstructured":"Goeller, M. et al. Epicardial adipose tissue density and volume are related to subclinical atherosclerosis, inflammation and major adverse cardiac events in asymptomatic subjects. J. Cardiovasc. Comput. Tomogr. 12, 67\u201373 (2018).","journal-title":"J. Cardiovasc. Comput. Tomogr."},{"key":"1020_CR7","doi-asserted-by":"publisher","first-page":"e190045","DOI":"10.1148\/ryai.2019190045","volume":"1","author":"F Commandeur","year":"2019","unstructured":"Commandeur, F. et al. Fully automated CT quantification of epicardial adipose tissue by deep learning: a multicenter study. Radio. Artif. Intell. 1, e190045 (2019).","journal-title":"Radio. Artif. Intell."},{"key":"1020_CR8","doi-asserted-by":"publisher","first-page":"e009829","DOI":"10.1161\/CIRCIMAGING.119.009829","volume":"13","author":"E Eisenberg","year":"2020","unstructured":"Eisenberg, E. et al. Deep learning\u2013based quantification of epicardial adipose tissue volume and attenuation predicts major adverse cardiovascular events in asymptomatic subjects. Circ. Cardiovasc. Imaging 13, e009829 (2020).","journal-title":"Circ. Cardiovasc. Imaging"},{"key":"1020_CR9","doi-asserted-by":"publisher","first-page":"76","DOI":"10.1016\/j.atherosclerosis.2020.11.008","volume":"318","author":"BK Tamarappoo","year":"2021","unstructured":"Tamarappoo, B. K. et al. Machine learning integration of circulating and imaging biomarkers for explainable patient-specific prediction of cardiac events: a prospective study. Atherosclerosis 318, 76\u201382 (2021).","journal-title":"Atherosclerosis"},{"key":"1020_CR10","doi-asserted-by":"publisher","first-page":"1194","DOI":"10.2967\/jnumed.118.220657","volume":"60","author":"PJ Slomka","year":"2019","unstructured":"Slomka, P. J., Miller, R. J. H., Hu, L. H., Germano, G. & Berman, D. S. Solid-state detector SPECT myocardial perfusion imaging. J. Nucl. Med. 60, 1194\u20131204 (2019).","journal-title":"J. Nucl. Med."},{"key":"1020_CR11","doi-asserted-by":"publisher","first-page":"652","DOI":"10.2967\/jnumed.122.264423","volume":"64","author":"RJ Miller","year":"2023","unstructured":"Miller, R. J. et al. Deep learning coronary artery calcium scores from SPECT\/CT attenuation maps improves prediction of major adverse cardiac events. J. Nucl. Med. 64, 652\u2013658 (2023).","journal-title":"J. Nucl. Med."},{"key":"1020_CR12","first-page":"100827","volume":"35","author":"C Trpkov","year":"2021","unstructured":"Trpkov, C. et al. Visually estimated coronary artery calcium score improves SPECT-MPI risk stratification. Int J. Cardiol. Heart Vasc. 35, 100827 (2021).","journal-title":"Int J. Cardiol. Heart Vasc."},{"key":"1020_CR13","doi-asserted-by":"publisher","first-page":"675","DOI":"10.1016\/j.jcmg.2022.06.006","volume":"16","author":"K Pieszko","year":"2023","unstructured":"Pieszko, K. et al. Deep learning of coronary calcium scores from PET\/CT attenuation maps accurately predicts adverse cardiovascular events. JACC Cardiovasc. Imaging 16, 675\u2013687 (2023).","journal-title":"JACC Cardiovasc. Imaging"},{"key":"1020_CR14","doi-asserted-by":"crossref","unstructured":"Isgum, I. et al. Automatic determination of cardiovascular risk by CT attenuation correction maps in Rb-82 PET\/CT. J. Nucl. Cardiol. 25, 2133\u201342 (2018).","DOI":"10.1007\/s12350-017-0866-3"},{"key":"1020_CR15","first-page":"774","volume":"13","author":"Y Otaki","year":"2020","unstructured":"Otaki, Y. et al. 5-year prognostic value of quantitative versus visual MPI in subtle perfusion defects. Results REFINE Spect. JACC Cardiovasc. Imaging 13, 774\u2013785 (2020).","journal-title":"Results REFINE Spect. JACC Cardiovasc. Imaging"},{"key":"1020_CR16","doi-asserted-by":"publisher","first-page":"1305","DOI":"10.1016\/j.jcmg.2011.10.003","volume":"4","author":"JM Bourque","year":"2011","unstructured":"Bourque, J. M. & Beller, G. A. Stress myocardial perfusion imaging for assessing prognosis: an update. JACC Cardiovasc. Imaging 4, 1305\u20131319 (2011).","journal-title":"JACC Cardiovasc. Imaging"},{"key":"1020_CR17","doi-asserted-by":"publisher","first-page":"567","DOI":"10.1093\/ehjci\/jez172","volume":"21","author":"RJH Miller","year":"2020","unstructured":"Miller, R. J. H. et al. Transient ischaemic dilation and post-stress wall motion abnormality increase risk in patients with less than moderate ischaemia. Eur. Heart J. Cardiovasc. Imaging 21, 567\u2013575 (2020).","journal-title":"Eur. Heart J. Cardiovasc. Imaging"},{"key":"1020_CR18","first-page":"416","volume":"4","author":"AH Talman","year":"2014","unstructured":"Talman, A. H. et al. Epicardial adipose tissue: far more than a fat depot. Cardiovasc. Diag Ther. 4, 416\u2013429 (2014).","journal-title":"Cardiovasc. Diag Ther."},{"key":"1020_CR19","doi-asserted-by":"publisher","first-page":"27","DOI":"10.1186\/s12933-021-01220-x","volume":"20","author":"A Lin","year":"2021","unstructured":"Lin, A. et al. Metabolic syndrome, fatty liver, and artificial intelligence-based epicardial adipose tissue measures predict long-term risk of cardiac events: a prospective study. Cardiovasc. Diabetol. 20, 27 (2021).","journal-title":"Cardiovasc. Diabetol."},{"key":"1020_CR20","doi-asserted-by":"publisher","first-page":"739","DOI":"10.1016\/j.numecd.2017.05.010","volume":"27","author":"V Camarena","year":"2017","unstructured":"Camarena, V. et al. Novel atherogenic pathways from the differential transcriptome analysis of diabetic epicardial adipose tissue. Nutr. Metab. Cardiovasc. Dis. 27, 739\u2013750 (2017).","journal-title":"Nutr. Metab. Cardiovasc. Dis."},{"key":"1020_CR21","first-page":"e18","volume":"101","author":"AG Bertaso","year":"2013","unstructured":"Bertaso, A. G., Bertol, D., Duncan, B. B. & Foppa, M. Epicardial fat: definition, measurements and systematic review of main outcomes. Arq. Bras. Cardiol. 101, e18\u2013e28 (2013).","journal-title":"Arq. Bras. Cardiol."},{"key":"1020_CR22","doi-asserted-by":"publisher","first-page":"416","DOI":"10.5152\/akd.2015.5991","volume":"15","author":"S Ero\u011flu","year":"2015","unstructured":"Ero\u011flu, S. How do we measure epicardial adipose tissue thickness by transthoracic echocardiography? Anatol. J. Cardiol. 15, 416\u2013419 (2015).","journal-title":"Anatol. J. Cardiol."},{"key":"1020_CR23","doi-asserted-by":"publisher","first-page":"154436","DOI":"10.1016\/j.metabol.2020.154436","volume":"115","author":"K Grodecki","year":"2021","unstructured":"Grodecki, K. et al. Epicardial adipose tissue is associated with extent of pneumonia and adverse outcomes in patients with COVID-19. Metabolism 115, 154436 (2021).","journal-title":"Metabolism"},{"key":"1020_CR24","doi-asserted-by":"publisher","first-page":"212","DOI":"10.1016\/j.atherosclerosis.2019.02.023","volume":"284","author":"G Iacobellis","year":"2019","unstructured":"Iacobellis, G. & Mahabadi, A. A. Is epicardial fat attenuation a novel marker of coronary inflammation? Atherosclerosis 284, 212\u2013213 (2019).","journal-title":"Atherosclerosis"},{"key":"1020_CR25","doi-asserted-by":"publisher","first-page":"2000","DOI":"10.1016\/j.jcmg.2018.11.032","volume":"12","author":"J Kwiecinski","year":"2019","unstructured":"Kwiecinski, J. et al. Peri-coronary adipose tissue density is associated with 18F-sodium fluoride coronary uptake in stable patients with high-risk plaques. JACC Cardiovasc. Imaging 12, 2000\u20132010 (2019).","journal-title":"JACC Cardiovasc. Imaging"},{"key":"1020_CR26","doi-asserted-by":"publisher","first-page":"8849429","DOI":"10.1155\/2021\/8849429","volume":"2021","author":"J Kwiecinski","year":"2021","unstructured":"Kwiecinski, J., Lassen, M. L. & Slomka, P. J. Advances in quantitative analysis of (18)F-sodium fluoride coronary imaging. Mol. Imaging 2021, 8849429 (2021).","journal-title":"Mol. Imaging"},{"key":"1020_CR27","doi-asserted-by":"publisher","first-page":"1075","DOI":"10.1007\/s12350-015-0370-6","volume":"24","author":"T Mazurek","year":"2017","unstructured":"Mazurek, T. et al. PET\/CT evaluation of 18 F-FDG uptake in pericoronary adipose tissue in patients with stable coronary artery disease: Independent predictor of atherosclerotic lesions\u2019 formation? J. Nucl. Cardiol. 24, 1075\u20131084 (2017).","journal-title":"J. Nucl. Cardiol."},{"key":"1020_CR28","doi-asserted-by":"publisher","first-page":"499","DOI":"10.3945\/ajcn.2008.27358","volume":"90","author":"J Ding","year":"2009","unstructured":"Ding, J. et al. The association of pericardial fat with incident coronary heart disease: the Multi-Ethnic Study of Atherosclerosis (MESA). Am. J. Clin. Nutr. 90, 499\u2013504 (2009).","journal-title":"Am. J. Clin. Nutr."},{"key":"1020_CR29","doi-asserted-by":"publisher","first-page":"1068","DOI":"10.1002\/oby.22798","volume":"28","author":"G Iacobellis","year":"2020","unstructured":"Iacobellis, G. & Gra-Menendez, S. Effects of dapagliflozin on epicardial fat thickness in patients with type 2 diabetes and obesity. Obesity 28, 1068\u20131074 (2020).","journal-title":"Obesity"},{"key":"1020_CR30","doi-asserted-by":"publisher","first-page":"749","DOI":"10.1161\/CIRCULATIONAHA.112.128413","volume":"127","author":"NJ Pagidipati","year":"2013","unstructured":"Pagidipati, N. J. & Gaziano, T. A. Estimating deaths from cardiovascular disease: a review of global methodologies of mortality measurement. Circulation 127, 749\u2013756 (2013).","journal-title":"Circulation"},{"key":"1020_CR31","doi-asserted-by":"publisher","first-page":"9","DOI":"10.1016\/j.ijcard.2020.04.012","volume":"313","author":"RJH Miller","year":"2020","unstructured":"Miller, R. J. H. et al. Prognostic significance of previous myocardial infarction and previous revascularization in patients undergoing SPECT MPI. Int J. Cardiol. 313, 9\u201315 (2020).","journal-title":"Int J. Cardiol."},{"key":"1020_CR32","first-page":"e344","volume":"130","author":"EA Amsterdam","year":"2014","unstructured":"Amsterdam, E. A. et al. AHA\/ACC guideline for the management of patients with non\u2013ST-elevation acute coronary syndromes. Circulation 130, e344\u2013e426 (2014). 2014.","journal-title":"Circulation"},{"key":"1020_CR33","doi-asserted-by":"publisher","first-page":"2020","DOI":"10.1161\/CIR.0b013e31826e1058","volume":"126","author":"K Thygesen","year":"2012","unstructured":"Thygesen, K. et al. Third universal definition of myocardial infarction. Circulation 126, 2020\u20132035 (2012).","journal-title":"Circulation"},{"key":"1020_CR34","doi-asserted-by":"publisher","first-page":"66","DOI":"10.1016\/j.nuclcard.2004.10.006","volume":"12","author":"PJ Slomka","year":"2005","unstructured":"Slomka, P. J. et al. Automated quantification of myocardial perfusion SPECT using simplified normal limits. J. Nucl. Cardiol. 12, 66\u201377 (2005).","journal-title":"J. Nucl. Cardiol."},{"key":"1020_CR35","doi-asserted-by":"crossref","unstructured":"Pieszko, K. et al. Calcium scoring in low-dose ungated chest CT scans using convolutional long-short term memory networks. Proc. SPIE 12032, 120323A (2022).","DOI":"10.1117\/12.2613147"},{"key":"1020_CR36","doi-asserted-by":"crossref","unstructured":"Shanbhag, A. D. et al. Comparative analysis between convolutional long short-term memory networks and vision transformers for coronary calcium scoring in non-contrast CT. Proc. SPIE 12464, 12464OD (2023).","DOI":"10.1117\/12.2655397"},{"key":"1020_CR37","doi-asserted-by":"publisher","first-page":"827","DOI":"10.1016\/0735-1097(90)90282-T","volume":"15","author":"AS Agatston","year":"1990","unstructured":"Agatston, A. S. et al. Quantification of coronary artery calcium using ultrafast computed tomography. J. Am. Coll. Cardiol. 15, 827\u2013832 (1990).","journal-title":"J. Am. Coll. Cardiol."}],"container-title":["npj Digital Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.nature.com\/articles\/s41746-024-01020-z.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41746-024-01020-z","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41746-024-01020-z.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,2,6]],"date-time":"2024-02-06T13:06:18Z","timestamp":1707224778000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.nature.com\/articles\/s41746-024-01020-z"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,2,3]]},"references-count":37,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2024,12]]}},"alternative-id":["1020"],"URL":"https:\/\/doi.org\/10.1038\/s41746-024-01020-z","relation":{},"ISSN":["2398-6352"],"issn-type":[{"value":"2398-6352","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,2,3]]},"assertion":[{"value":"6 September 2023","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"18 January 2024","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"3 February 2024","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"Dr. Robert Miller received consulting fees and research support from Pfizer. Drs. Berman, Slomka and Mr. Paul B. Kavanagh participate in software royalties for QPS software at Cedars-Sinai Medical Center. Dr Slomka has received consulting fees from Synektik. Dr. Berman has served as a consultant for GE Healthcare. The remaining authors declare no competing interests.","order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"24"}}