Abstract
BackgroundTranscatheter aortic valve implantation (TAVI) has become a commonly applied procedure for high-risk aortic valve stenosis patients. However, for some patients, this procedure does not result in the expected benefits. Previous studies indicated that it is difficult to predict the beneficial effects for specific patients. We aim to study the accuracy of various traditional machine learning (ML) algorithms in the prediction of TAVI outcomes.Methods and resultsClinical and laboratory data from 1,478 TAVI patients from a single centre were collected. The outcome measures were improvement of dyspnoea and mortality. Three experiments were performed using (1) screening data, (2) laboratory data, and (3) the combination of both. Five well-established ML techniques were implemented, and the models were evaluated based on the area under the curve (AUC). Random forest classifier achieved the highest AUC (0.70) for predicting mortality. Logistic regression had the highest AUC (0.56) in predicting improvement of dyspnoea.ConclusionsIn our single-centre TAVI population, the tree-based models were slightly more accurate than others in predicting mortality. However, ML models performed poorly in predicting improvement of dyspnoea.Electronic supplementary materialThe online version of this article (10.1007/s12471-019-1285-7) contains supplementary material, which is available to authorized users.
Highlights
Aortic valve stenosis (AS) is one of the most common valvular heart diseases, impacting, in general, the elderly population
To elucidate which features may be important in the machine learning (ML) techniques, the average feature importance for random forest classifier (RFC) and gradient tree boosting (GTB) was calculated based on the number of times the feature was selected for splitting and weighted by the average squared improvement of the model over all trees [20]
The model based on RFC was most accurate with an area under the curve (AUC) of 0.70 [Q1 0.67 – Q3 0.74] and the results are considered to be significantly different according to the Wilcoxon test
Summary
Aortic valve stenosis (AS) is one of the most common valvular heart diseases, impacting, in general, the elderly population. Transcatheter aortic valve implantation (TAVI) has developed into a routine treatment for AS patients at elevated risk of surgery. There is strict patient selection for the TAVI procedure and various planning and treatment support tools are available [1,2,3], a number of patients have limited benefit from TAVI [4]. Current risk models have only limited accuracy in predicting TAVI outcomes [5]. Transcatheter aortic valve implantation (TAVI) has become a commonly applied procedure for high-risk aortic valve stenosis patients. We aim to study the accuracy of various traditional machine learning (ML) algorithms in the prediction of TAVI outcomes
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: Netherlands heart journal : monthly journal of the Netherlands Society of Cardiology and the Netherlands Heart Foundation
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
