Role of AI-Based Echocardiography in Early Detection of Subclinical Heart Failure
Article Sidebar
Main Article Content
Abstract
Background: Subclinical left ventricular dysfunction may precede symptomatic heart failure, while conventional echocardiographic interpretation can be affected by operator dependence and measurement variability. Artificial intelligence (AI)-assisted echocardiography may facilitate more standardized recognition of subtle abnormalities in ventricular function. Objective: To evaluate the diagnostic accuracy of an AI-based echocardiography model for detecting subclinical left ventricular dysfunction compared with conventional echocardiographic assessment. Methods: This diagnostic accuracy study included 212 adults undergoing echocardiographic evaluation at a cardiac care facility in Faisalabad, Pakistan. Standard transthoracic echocardiography served as the reference assessment, and stored echocardiographic images and video clips were independently analyzed using an AI-based model. Diagnostic performance was assessed using sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), overall accuracy, and receiver operating characteristic analysis. Results: Conventional echocardiography identified subclinical left ventricular dysfunction in 58 of 212 participants (27.4%). Based on the internally reconciled classification counts, the AI model correctly identified 54 participants with dysfunction and 146 without dysfunction, with 4 false-negative and 8 false-positive classifications. Sensitivity was 93.1%, specificity 94.8%, PPV 87.1%, NPV 97.3%, and overall accuracy 94.3%. The area under the receiver operating characteristic curve was 0.95 (95% CI: 0.91–0.98), indicating excellent discrimination. Conclusion: AI-based echocardiography demonstrated high diagnostic accuracy for identifying subclinical left ventricular dysfunction in this single-center sample. The findings support its potential as an adjunct to conventional echocardiographic interpretation, while broader validation is required before routine clinical implementation
Article Details
Issue
Section
This work is licensed under a Creative Commons Attribution 4.0 International License.
How to Cite
References
1. Lang RM, Badano LP, Mor-Avi V, Afilalo J, Armstrong A, Ernande L, et al. Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. Eur Heart J Cardiovasc Imaging. 2015;16(3):233–270. doi:10.1093/ehjci/jev014.
2. Duffy G, Cheng PP, Yuan N, He B, Kwan AC, Shun-Shin MJ, et al. High-throughput precision phenotyping of left ventricular hypertrophy with cardiovascular deep learning. JAMA Cardiol. 2022;7(4):386–395.
3. Ouyang D, He B, Ghorbani A, Yuan N, Ebinger J, Langlotz CP, et al. Video-based AI for beat-to-beat assessment of cardiac function. Nature. 2020;580:252–256. doi:10.1038/s41586-020-2140-2.
4. Ghorbani A, Ouyang D, Abid A, He B, Chen JH, Harrington RA, et al. Deep learning interpretation of echocardiograms. NPJ Digit Med. 2020;3:10. doi:10.1038/s41746-020-0235-9.
5. Zhang J, Gajjala S, Agrawal P, Tison GH, Hallock LA, Beussink-Nelson L, et al. Fully automated echocardiogram interpretation in clinical practice. Circulation. 2018;138(16):1623–1635. doi:10.1161/CIRCULATIONAHA.118.034338.
6. Sengupta PP, Shrestha S, Berthon B, Messas E, Donal E, Tison GH, et al. Artificial intelligence and machine learning in echocardiography: an evolving landscape. JACC Cardiovasc Imaging. 2021;14(11):2207–2218. doi:10.1016/j.jcmg.2021.07.004.
7. Khan MS, Siddiqi TJ, Khan SU, Shah SJ, Van Spall HGC, Greene SJ, et al. Artificial intelligence in cardiovascular imaging: current applications and future directions. Eur Heart J Digit Health. 2021;2(3):395–406. doi:10.1093/ehjdh/ztab045.
8. Jafar TH, Qadri Z, Chaturvedi N. Coronary artery disease epidemic in Pakistan: challenges and opportunities. Heart Asia. 2009;1:10–14. doi:10.1136/ha.2008.000331.
9. Aziz S, Noor L, Fahim M, et al. Cardiovascular disease risk factors in Pakistani population. Pak J Med Sci. 2020;36(4):745–750. doi:10.12669/pjms.36.4.1812.
10. Khan MA, Hashmi S, Khan MS, et al. Burden of cardiovascular diseases in Pakistan: epidemiological review. J Pak Med Assoc. 2021;71(2):S1–S6.
11. Bossuyt PM, Reitsma JB, Bruns DE, Gatsonis CA, Glasziou PP, Irwig L, et al. STARD 2015: an updated list of essential items for reporting diagnostic accuracy studies. BMJ. 2015;351:h5527. doi:10.1136/bmj.h5527.
12. Narula S, Shameer K, Salem Omar AM, Dudley JT, Sengupta PP. Machine-learning algorithms to automate morphological and functional assessments in 2D echocardiography. J Am Coll Cardiol. 2016;68(21):2287–2295. doi:10.1016/j.jacc.2016.08.062.
13. Madani A, Ong JR, Tibrewal A, Mofrad MRK. Deep echocardiography: data-efficient supervised and semi-supervised deep learning towards automated diagnosis of cardiac disease. NPJ Digit Med. 2018;1:59.
14. Narang A, Bae R, Hong H, Thomas Y, Surette S, Cadieu C, et al. Utility of a deep-learning algorithm to guide novices to acquire echocardiograms for limited diagnostic use. JAMA Cardiol. 2021;6(6):624–632. doi:10.1001/jamacardio.2021.0618.
15. Krittanawong C, Zhang H, Wang Z, Aydar M, Kitai T. Artificial intelligence in precision cardiovascular medicine. J Am Coll Cardiol. 2017;69(21):2657–2664. doi:10.1016/j.jacc.2017.03.571.
16. Topol EJ. High-performance medicine: the convergence of human and artificial intelligence. Nat Med. 2019;25:44–56. doi:10.1038/s41591-018-0300-7.
17. Kwan AC, Cohen JA, Rebello MS, et al. Machine learning for cardiovascular imaging. JACC Cardiovasc Imaging. 2022;15(8):1397–1410. doi:10.1016/j.jcmg.2022.02.012.
18. Sengupta PP, Kulkarni H, Narula J. Artificial intelligence in echocardiography: clinical translation. JACC Cardiovasc Imaging. 2022;15(5):876–889. doi:10.1016/j.jcmg.2021.12.017.
19. Kusunose K, Haga A, Abe T, Sata M. Utilization of artificial intelligence in echocardiography. J Cardiol. 2020;76(3):227–233. doi:10.1016/j.jjcc.2019.12.009.