Prediction of Coronary Artery Calcium and Cardiovascular Risk on Chest Radiographs Using Deep Learning
Peter I Kamel, Paul H Yi, Haris I Sair, Cheng Ting Lin, Peter I Kamel, Paul H Yi, Haris I Sair, Cheng Ting Lin
Abstract
Purpose: To assess the ability of deep convolutional neural networks (DCNNs) to predict coronary artery calcium (CAC) and cardiovascular risk on chest radiographs.
Materials and methods: In this retrospective study, 1689 radiographs in patients who underwent cardiac CT and chest radiography within the same year, between 2013 and 2018, were included (mean age, 56 years ± 11 [standard deviation]; 969 radiographs in women). Agatston scores were used as ground truth labels for DCNN training on radiographs. DCNNs were trained for binary classification of (a) nonzero or zero total calcium scores, (b) presence or absence of calcium in each coronary artery, and (c) total calcium scores above or below varying thresholds. Results from classification of test images were compared with established 10-year atherosclerotic cardiovascular disease (ASCVD) risk scores in each cohort. Classifier performance was measured using area under the receiver operating characteristic curve (AUC) with attention maps to highlight areas of decision-making.
Results: Binary classification between zero and nonzero total calcium scores reached an AUC of 0.73 on frontal radiographs, with similar performance on laterals (AUC, 0.70; P = .56). Performance was similar for binary classification of absolute total calcium score above or below 100 (AUC, 0.74). Frontal radiographs that tested positive for a predicted nonzero CAC score correlated with a higher 10-year ASCVD risk of 17.2% ± 10.9 compared with 11.9% ± 10.2 for a negative test, indicating predicted CAC score of zero (P < .001). Multivariate logistic regression demonstrated the algorithm could predict a nonzero calcium score independent of traditional cardiovascular risk factors. Performance was reduced for individual coronary arteries. Heat maps primarily localized to the cardiac silhouette and occasionally other cardiovascular findings.
Conclusion: DCNNs trained on chest radiographs had modest accuracy for predicting the presence of CAC correlating with cardiovascular risk.Keywords: Coronary Arteries, Cardiac, Calcifications/Calculi, Neural NetworksSee also the commentary by Gupta and Blankstein in this issue.©RSNA, 2021.
Keywords: Calcifications/Calculi; Cardiac; Coronary Arteries; Neural Networks.
Conflict of interest statement
Disclosures of Conflicts of Interest: P.I.K. disclosed no relevant relationships. P.H.Y. disclosed no relevant relationships. H.I.S. Activities related to the present article: disclosed no relevant relationships. Activities not related to the present article: author received payment from Demler, Armstrong, and Rowland for defense expert testimony; author's institution has grants/grants pending from the National Institutes of Health. Other relationships: disclosed no relevant relationships. C.T.L. disclosed no relevant relationships.
2021 by the Radiological Society of North America, Inc.
Figures
References
- Parikh P, Shah N, Ahmed H, Schoenhagen P, Fares M. Coronary artery calcium scoring: Its practicality and clinical utility in primary care. Cleve Clin J Med 2018;85(9):707–716.
- McClelland RL, Chung H, Detrano R, Post W, Kronmal RA. Distribution of coronary artery calcium by race, gender, and age: results from the Multi-Ethnic Study of Atherosclerosis (MESA). Circulation 2006;113(1):30–37.
- Kondos GT, Hoff JA, Sevrukov A, et al. . Electron-beam tomography coronary artery calcium and cardiac events: a 37-month follow-up of 5635 initially asymptomatic low- to intermediate-risk adults. Circulation 2003;107(20):2571–2576.
- Wilson PW, D’Agostino RB, Levy D, Belanger AM, Silbershatz H, Kannel WB. Prediction of coronary heart disease using risk factor categories. Circulation 1998;97(18):1837–1847.
- Williams KA Sr, Kim JT, Holohan KM. Frequency of unrecognized, unreported, or underreported coronary artery and cardiovascular calcification on noncardiac chest CT. J Cardiovasc Comput Tomogr 2013;7(3):167–172.
- Margolis JR, Chen JT, Kong Y, Peter RH, Behar VS, Kisslo JA. The diagnostic and prognostic significance of coronary artery calcification. A report of 800 cases. Radiology 1980;137(3):609–616.
- Sakuma H, Takeda K, Hirano T, et al. . Plain chest radiograph with computed radiography: improved sensitivity for the detection of coronary artery calcification. AJR Am J Roentgenol 1988;151(1):27–30.
- Agatston AS, Janowitz WR, Hildner FJ, Zusmer NR, Viamonte M Jr, Detrano R. Quantification of coronary artery calcium using ultrafast computed tomography. J Am Coll Cardiol 1990;15(4):827–832.
- Hecht HS, Cronin P, Blaha MJ, et al. . 2016SCCT/STR guidelines for coronary artery calcium scoring of noncontrast noncardiac chest CT scans: A report of the Society of Cardiovascular Computed Tomography and Society of Thoracic Radiology. J Cardiovasc Comput Tomogr 2017;11(1):74–84 [Published correction appears in J Cardiovasc Comput Tomogr 2017;11(2):170.].
- Halabi SS, Prevedello LM, Kalpathy-Cramer J, et al. . The RSNA Pediatric Bone Age Machine Learning Challenge. Radiology 2019;290(2):498–503.
- Istenes B. Python ASCVD. . Published 2018. Accessed November 10, 2020.
- The Multi-Ethnic Study of Atherosclerosis. . Published 2020. Accessed November 10, 2020.
- Mader S. Attention on Pretrained-VGG16 for Bone Age. . Published 2018. Accessed November 10, 2020.
- Pedregosa F, Varoquaux G, Gramfort A, et al. . Scikit-learn: Machine Learning in Python. JMLR 12, pp. 2825–2830, 2011. . Accessed November 10, 2020.
- Wolterink JM, Leiner T, Takx RAP, Viergever MA, Isgum I. Automatic Coronary Calcium Scoring in Non-Contrast-Enhanced ECG-Triggered Cardiac CT With Ambiguity Detection. IEEE Trans Med Imaging 2015;34(9):1867–1878.
- Shadmi R, Mazo V, Bregman-Amitai O, Elnekave E. Fully-convolutional deep-learning based system for coronary calcium score prediction from non-contrast chest CT. In: 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018), Washington, DC, April 4–7, 2018. Piscataway, NJ: IEEE, 2018; 24–28.
- Cano-Espinosa C, González G, Washko GR, Cazorla M, Estépar RSJ. Automated Agatston score computation in non-ECG gated CT scans using deep learning. In: Angelini ED, Landman BA, eds. Proceedings of SPIE: medical imaging 2018—image processing. Vol 10574. Bellingham, Wash: International Society for Optics and Photonics, 2018; 105742K
- Hollander J, Chase M. Evaluation of the adult with chest pain in the emergency department. UpToDate Web site. . Published 2019. Accessed November 10, 2020.
- Greenland P, Bonow RO, Brundage BH, et al. . ACCF/AHA 2007 clinical expert consensus document on coronary artery calcium scoring by computed tomography in global cardiovascular risk assessment and in evaluation of patients with chest pain: a report of the American College of Cardiology Foundation Clinical Expert Consensus Task Force (ACCF/AHA Writing Committee to Update the 2000 Expert Consensus Document on Electron Beam Computed Tomography) developed in collaboration with the Society of Atherosclerosis Imaging and Prevention and the Society of Cardiovascular Computed Tomography. J Am Coll Cardiol 2007;49(3):378–402.
- Iijima K, Hashimoto H, Hashimoto M, et al. . Aortic arch calcification detectable on chest X-ray is a strong independent predictor of cardiovascular events beyond traditional risk factors. Atherosclerosis 2010;210(1):137–144.
- Atar S, Jeon DS, Luo H, Siegel RJ. Mitral annular calcification: a marker of severe coronary artery disease in patients under 65 years old. Heart 2003;89(2):161–164.
Source: PubMed