Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2019 Jul 22;12(14):1328-1338.
doi: 10.1016/j.jcin.2019.06.013.

Machine Learning Prediction Models for In-Hospital Mortality After Transcatheter Aortic Valve Replacement

Dagmar F Hernandez-Suarez  1 , Yeunjung Kim  2 , Pedro Villablanca  3 , Tanush Gupta  4 , Jose Wiley  4 , Brenda G Nieves-Rodriguez  5 , Jovaniel Rodriguez-Maldonado  5 , Roberto Feliu Maldonado  5 , Istoni da Luz Sant'Ana  5 , Cristina Sanina  4 , Pedro Cox-Alomar  6 , Harish Ramakrishna  7 , Angel Lopez-Candales  8 , William W O'Neill  3 , Duane S Pinto  9 , Azeem Latib  4 , Abiel Roche-Lima  5
Affiliations

Machine Learning Prediction Models for In-Hospital Mortality After Transcatheter Aortic Valve Replacement

Dagmar F Hernandez-Suarez et al. JACC Cardiovasc Interv. .

Abstract

Objectives: This study sought to develop and compare an array of machine learning methods to predict in-hospital mortality after transcatheter aortic valve replacement (TAVR) in the United States.

Background: Existing risk prediction tools for in-hospital complications in patients undergoing TAVR have been designed using statistical modeling approaches and have certain limitations.

Methods: Patient data were obtained from the National Inpatient Sample database from 2012 to 2015. The data were randomly divided into a development cohort (n = 7,615) and a validation cohort (n = 3,268). Logistic regression, artificial neural network, naive Bayes, and random forest machine learning algorithms were applied to obtain in-hospital mortality prediction models.

Results: A total of 10,883 TAVRs were analyzed in our study. The overall in-hospital mortality was 3.6%. Overall, prediction models' performance measured by area under the curve were good (>0.80). The best model was obtained by logistic regression (area under the curve: 0.92; 95% confidence interval: 0.89 to 0.95). Most obtained models plateaued after introducing 10 variables. Acute kidney injury was the main predictor of in-hospital mortality ranked with the highest mean importance in all the models. The National Inpatient Sample TAVR score showed the best discrimination among available TAVR prediction scores.

Conclusions: Machine learning methods can generate robust models to predict in-hospital mortality for TAVR. The National Inpatient Sample TAVR score should be considered for prognosis and shared decision making in TAVR patients.

Keywords: machine learning; mortality; transcatheter aortic valve replacement.

PubMed Disclaimer

Figures

Figure 1.
Figure 1.
Operating characteristic area under the curve (AUC) by machine learning model.LR, Logistic Regression; ANN, Artificial Neural Network; NB, Naïve Bayes; RF, Random Forest.
Figure 2.
Figure 2.
Predictive performance of TAVR in-hospital mortality models by number of variables. LR, Logistic Regression; ANN, Artificial Neural Network; NB, Naïve Bayes; RF, Random Forest.
Figure 3.
Figure 3.
Machine learning model-based feature importance according to the mean ranking. * LR, Logistic Regression; ANN, Artificial Neural Network; NB, Naïve Bayes; RF, Random Forest. *The variable’s importance is inversely proportional to the rank. The X-axis depicts the top 10 variables ordered according to mean ranking.
Central Illustration.
Central Illustration.
Overview of the methods used for data extraction, training and testing. Data from the NIS database was extracted using a tailored scripting in Python to select patients with TAVR. Then the cohort was split into a training set (70% of the data, n=7,615) and test set (30% of the data, n=3,268). A feature ranking method was applied to the training set to determine the top 5,10,15,20,30 variables. Because of data imbalance the training sets were randomly oversampled and then the ML algorithms (i.e. LR, ANN, NB, RF) were trained to develop the models. The different set of variables (including All variables = 43) were used independently to train each of the ML algorithms. The developed models were validated using the test sets and computing the precision metric results focused on the AUC. NIS, National Inpatient Sample; TAVR, Transcatheter Aortic Valve Replacement; ML, Machine Learning; ANN, Artificial Neural Network; LR, Logistic Regression; NB, Naïve Bayes; RF, Random Forest; AUC, area under the curve..
See this image and copyright information in PMC

Comment in

References

    1. Nishimura RA, Otto CM, Bonow RO, et al. 2017 AHA/ACC Focused Update of the 2014 AHA/ACC Guideline for the Management of Patients With Valvular Heart Disease: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. J Am Coll Cardiol. 2017;70(2):252–289. - PubMed
    1. Chamandi C, Puri R, Rodriguez-Gabella T, Rodés-Cabau J. Latest generation transcatheter aortic valve replacement devices and procedures. Can J Cardiol 2017;33:1082–90. - PubMed
    1. Mack MJ, Leon MB, Thourani VH, et al. Transcatheter Aortic-Valve Replacement with a Balloon-Expandable Valve in Low-Risk Patients. N Engl J Med 2019;380:1695–1705. - PubMed
    1. TAVR In-Hospital Mortality Risk Calculator. https://tools.acc.org/tavrrisk/#!/content/evaluate/. Accessed February 28, 2018.
    1. Edwards FH, Cohen DJ, O’Brien SM, et al. Development and Validation of a Risk Prediction Model for In-Hospital Mortality After Transcatheter Aortic Valve Replacement. JAMA Cardiol 2016;1(1):46–52. - PubMed

Publication types