Desarrollo de un modelo por inteligencia artificial con hemodinamia no invasiva para predecir preeclampsia en embarazos de alto riesgo
Revista Argentina de Cardiología · 31.10.2023
DOI: 10.7775/rac.v91.i5.20665 ↗Kurz gefasst
Fachartikel in der Zeitschrift Revista Argentina de Cardiología, online erschienen am 31.10.2023. Verfasst von Ricardo D. Olano und 9 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-SA 4.0).
Abstract (Originalsprache)
Background
Preeclampsia (PE) is the main cause of maternal-fetal morbidity and mortality in our country. Early hemodynamic changes during pregnancy could predict progression to PE. Machine learning (ML) enables the discovery of hidden patterns that could early detect PE development.
Objective
The aim of this study was to build a classification tree with non-invasive hemodynamic variables for the early prediction of PE occurrence.
Methods
This was a prospective observational study with high-risk pregnant women (n=1155) referred by the Obstetrics division from January 2016 to October 2022 for ML training sampling, with a j48 classification tree. A total of 112 women with 10 to 16-week pregnancy and without pharmacological treatment were selected, who completed follow-up at the end of their pregnancy. The end point, PE, was a composite of preeclampsia, eclampsia, and HELLP syndrome. They were evaluated simultaneously with impedance cardiography and pulse wave velocity and with 24-h ambulatory blood pressure monitoring (ABPM).
Results
Seventeen patients (15.18%) presented PE. A predictive classification tree was generated with arterial compliance index (ACI), cardiac index (CI), cardiac work index (CWI), ejective time ratio (ETR), and Heather index (HI). A total of 93.75% patients were correctly classified (Kappa 0.70, positive predictive value 0.94 and negative predictive value 0.35; accuracy 0.94, and area under the ROC curve 0.93).
Conclusion
ACI, CI, CWI, ETR and HI variables predicted the early development of PE in our sample with excellent discrimination and accuracy, non-invasively, safely and at low cost. Keywords: Machine Learning - Preeclampsia - Non-invasive Hemodynamics - Impedance Cardiography How to cite this article Olano RD, Espeche W, Leiva Sisnieguez BC, Carrera Ramos PM, Martínez C, Leiva Sisnieguez CE, et al. Artificial Intelligence Modeling with Non-Invasive Hemodynamics to Predict Preeclampsia in High-Risk Pregnancy. Rev Argent Cardiol 2023;91:330-35. http://dx.doi.org/10.7775/rac.v91.i5.20675
Zitieren
APA
Olano, R. D., Espeche, W. G., Sisnieguez, B. C. L., Ramos, P. M. C., Martínez, C. M., Sisnieguez, C. E. L., Iraola, A. D., Gómez, D. R., Minetto, J., & Salazar, M. R. (2023). Desarrollo de un modelo por inteligencia artificial con hemodinamia no invasiva para predecir preeclampsia en embarazos de alto riesgo. Revista Argentina de Cardiología. https://doi.org/10.7775/rac.v91.i5.20665
Vancouver
Olano RD, Espeche WG, Sisnieguez BCL, Ramos PMC, Martínez CM, Sisnieguez CEL, et al. Desarrollo de un modelo por inteligencia artificial con hemodinamia no invasiva para predecir preeclampsia en embarazos de alto riesgo. Revista Argentina de Cardiología. 2023. doi:10.7775/rac.v91.i5.20665
BibTeX
@article{Olano2023,
author = {Olano, Ricardo D. and Espeche, Walter G. and Sisnieguez, Betty C. Leiva and Ramos, Patricia M . Carrera and Martínez, Camilo Martínez and Sisnieguez, Carlos E. Leiva and Iraola, Ana De and Gómez, Daniela R. and Minetto, Julián and Salazar, Martín R.},
title = {Desarrollo de un modelo por inteligencia artificial con hemodinamia no invasiva para predecir preeclampsia en embarazos de alto riesgo},
journal = {Revista Argentina de Cardiología},
year = {2023},
doi = {10.7775/rac.v91.i5.20665}
}
EUVIMED