What does this research mean for the field?

An artificial intelligence model predicts the progression from mild to severe aortic stenosis with an AUC-ROC of 0.93 and an accuracy of 85% using echocardiographic data. Novelty: ClaimNovelty.NOVEL_FINDING. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

This study aims to develop a machine learning algorithm for predicting the progression of aortic stenosis using echocardiographic data.

February 8, 2026

Prediction of aortic stenosis progression using artificial intelligence: a machine learning model based on echocardiographic data

Key Result

An AI model using echocardiographic data predicted progression to severe aortic stenosis with 85% accuracy and 0.93 AUC-ROC in 9,330 patients.

Key Points

This study aims to develop a machine learning algorithm for predicting the progression of aortic stenosis using echocardiographic data.
Retrospective analysis of 9,330 echocardiograms from patients with mild or moderate AS
Utilization of the CatBoost algorithm for model development
Assessment of model performance through accuracy, AUC-ROC, and calibration
Interpretability provided via SHAP values
Model achieved an AUC-ROC of 0.93 and accuracy of 85%
Sensitivity of the model was recorded at 86%
Key predictors included aortic valve gradients and aortic diameter
Model demonstrated robust calibration and generalizability through cross-validation

Structured PICO

Does a machine learning algorithm based on echocardiographic data accurately predict the progression of mild or moderate aortic stenosis to severe stages?

Population

Patients originally diagnosed with mild or moderate aortic stenosis (AS) from a single center database (n=9,330 echocardiograms)

Intervention

CatBoost machine learning algorithm based on echocardiographic data

Outcome

Progression of aortic stenosis (AS) from mild to severe stagessurrogate

An echocardiography-focused machine learning model can accurately predict the progression of mild or moderate aortic stenosis to severe stages, potentially facilitating personalized follow-up strategies.

Main Result

Absolute Event Rate: 0% vs 0%

Abstract

Abstract Introduction Aortic Stenosis (AS) is amongst the most common valvular heart diseases, particularly in older individuals. Left untreated, the condition carries a significant risk of morbidity and mortality. Current guidelines for monitoring AS progression focus on serial echocardiographic assessment, which is resource-intensive and subject to variability. Artificial intelligence (AI) may offer an opportunity to enhance the early identification of patients at risk of developing severe AS. Purpose The objective of this study is to develop and validate a machine learning algorithm that can accurately predict the progression of aortic stenosis (AS) from mild to severe stages using echocardiographic data. Methods We retrospectively analyzed a single center database of 9,330 echocardiograms performed on patients originally diagnosed with mild or moderte AS and followed for up to 5 years. Our model was developed using the CatBoost algorithm, which is agnostic to any patient data outside the scope of the echocardiography report. Performance was assessed for accuracy, AUC-ROC, and calibration. SHAP values provided interpretability for the model’s predictions. Results The model demonstrated a strong predictive performance, manifested by an AUC-ROC of 0.93, an accuracy of 85%, and a sensitivity of 86%. Key predictors for severe AS development included maximal and mean systolic aortic valve gradients, calculated aortic valve area, and ascending aortic diameter. The model successfully identified patients at high risk of progression, with robust calibration and generalizability confirmed through cross-validation. Conclusions Our novel, echocardiography-focused AI model is a reliable tool for the early identification of patients at risk of progression to severe AS. Pending future, multi-center, prospective validation, such models may facilitate personalized follow-up strategies and timely interventions, ultimately leading to improved patient outcomes resource utilization.

Bookmark