Machine learning approaches for risk prediction after percutaneous coronary intervention - a systematic review and meta-analysis

Aug 2, 2026

Upload by Riley

Accurate prediction of clinical outcomes after percutaneous coronary intervention (PCI) is critical for risk mitigation and peri-procedural planning. Traditional risk models show limited predictive performance, while machine learning (ML) may enable improved risk stratification. This systematic review and meta-analysis compared ML models with traditional statistical methods using C-statistics across mortality, major bleeding, and MACE. Thirty-four models from 13 observational studies were synthesized, showing marginal ML advantages for MACE and major bleeding, with limited external validation.

LanguageEnglish

CategoryHealthcare

FormatPDF

Pages23

Reading Time58 min

AccessFree to download

Aims

Methods and results

Outcomes and comparative discrimination

Included studies and validation

Conclusion

Keywords

Machine learning approaches for risk prediction after percutaneous coronary intervention: a systematic review and meta-analysis Author Zaka , Ammar, Mutahar, Daud , Gorcilov, James , Gupta , Aashray K , Kovoor, Joshua G , Stretton , Brandon , Mridha , Naim , Sivagangabalan , Gopal , Thiagalingam , Aravinda , Chow, Clara K , Zaman , Sarah , Jayasinghe , Rohan , Kovoor, Pramesh , Bacchi , Stephen Published 2024 Journal Title European Heart Journal -Digital Health Version Version of Record (VoR) DOI 10.1093/ehjdh/ztae074 Rights statement © The Author(s) 2024. Published by Oxford University Press on behalf of the European Society of Cardiology. This is an Open Access article distributed under the terms of the Creative Commons Attribution License ([https://creativecommons.org/licenses/by/4.0/](https://creativecommons.org/licenses/by/4.0/)), which permits unrestricted reuse , distribution , and reproduction in any medium , provided the original work is properly cited. Downloaded from [https://hdl.handle.net/10072/434355](https://hdl.handle.net/10072/434355) Griffith Research Online [https://research-repository.griffith.edu.au](https://research-repository.griffith.edu.au) European Heart Journal-Digital Health (2024) 00, 1–22 ORIGINAL ARTICLE [https://doi.org/10.1093/ehjdh/ztae074](https://doi.org/10.1093/ehjdh/ztae074 Artificial intelligence)[ Artificial intelligence](https://doi.org/10.1093/ehjdh/ztae074 Artificial intelligence) (machine learning, deep learning) Machine learning approaches for risk prediction after percutaneous coronary intervention: a systematic review and meta-analysis Ammar Zaka 1,*, Daud Mutahar2, James Gorcilov2, Aashray K. Gupta3,4, Joshua G. Kovoor3,5, Brandon Stretton3, Naim Mridha6, Gopal Sivagangabalan7,8, Aravinda Thiagalingam8,9, Clara K. Chow 8,9, Sarah Zaman 8,9, Rohan Jayasinghe1, Pramesh Kovoor 8,9, and Stephen Bacchi10 1Department of Cardiology, Gold Coast University Hospital, 1 Hospital Boulevard, Southport, QLD 4215, Australia; 2Faculty of Health Sciences and Medicine, Bond University, 14 University Drive, Robina, QLD 4216, Australia; 3University of Adelaide, Adelaide, SA 5005, Australia; 4Royal North Shore Hospital, Reserve Rd, St Leonards, NSW 2065, Australia; 5Ballarat Base Hospital, 1 Drummond St N, Ballarat Central, VIC 3350, Australia; 6Department of Cardiology, The Prince Charles Hospital, 627 Rode Rd, Chermside, QLD 4032, Australia; 7University of Notre Dame, 128-140 Broadway, Chippendale, NSW 2007, Australia; 8Department of Cardiology, Westmead Hospital, Cnr Hawkesbury Road and Darcy Rd, Westmead, NSW 2145, Australia; 9Faculty of Medicine and Health, Westmead Applied Research Centre, University of Sydney, NSW, Australia; and 10Massachusetts General Hospital, 55 Fruit St, Boston, MA 02114, USA Received 8 June 2024; revised 30 July 2024; accepted 23 September 2024; online publish-ahead-of-print 14 October 2024 Aims Accurate prediction of clinical outcomes following percutaneous coronary intervention (PCI) is essential for mitigating risk and peri-procedural planning. Traditional risk models have demonstrated a modest predictive value. Machine learning (ML) models offer an alternative risk stratification that may provide improved predictive accuracy. Methods and results This study was reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses, Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modelling Studies and Transparent Reporting of a multivariable prediction model for Individual Prognosis or Diagnosis guidelines. PubMed, EMBASE, Web of Science, and Cochrane databases were searched until 1 November 2023 for studies comparing ML models with traditional statistical methods for event prediction after PCI. The primary outcome was comparative discrimination measured by C-statistics with 95% confidence intervals (CIs) between ML models and traditional methods in estimating th

FAQs

What clinical problem does the review address after PCI?

The review focuses on predicting clinical outcomes following percutaneous coronary intervention to support risk mitigation and peri-procedural planning.

How were machine learning models compared with traditional risk models?

Studies comparing ML models with traditional statistical methods were pooled and evaluated primarily by discrimination using C-statistics with 95% confidence intervals for mortality, major bleeding, and MACE.

What were the main findings regarding ML performance versus traditional methods?

ML models marginally outperformed traditional methods for discrimination of MACE and major bleeding after PCI, while results for all-cause mortality did not show a meaningful advantage.

Aug 2, 2026

Upload by Riley