AI-driven multimodal clinical decision support for diagnostic of the atrial cardiomyopathy and the risk of atrial fibrillation
M Ponnaiah, P Van Hille, D Filgueiras Rama, L Soulat-Dufour, N Kachenour, C Antoniades, U Schotten, D Dobrev, L Fabritz, A Cohen, A Goette, S N HatemAbstract
Background
Early and personalized diagnostic of the atrial cardiomyopathy that paves the way of atrial fibrillation (AF) is necessary for timely deployment of prevention strategies and personalised medical care. However, the diagnostic of the atrial cardiomyopathy needs to integrate multiple parameters including imaging, electrocardiographic, biological, genomic.
Purpose
To develop and validate an interoperable demonstrator integrating imaging, electrophysiology, clinical and omics data using explainable AI to support personalized diagnostic of the atrial cardiomyopathy and of the risk prediction of AF.
Methods
We harmonized multimodal datasets across several European centers, including clinical variables, ECG, echocardiography, cardiac-MRI, CT, wearable-datasets and omics. Predictive models were developed using ensemble machine learning algorithms (e.g., CatBoost, XGBoost), deep learning approaches for imaging segmentation, and multi-block data integration frameworks. Model interpretability was addressed through SHAP value analysis and partial dependence plots. The demonstrator was built as a GDPR-compliant, decision support tool with clinician-friendly interfaces.
Results
The demonstrator integrates validated models predicting AF recurrence, stroke risk, and atrial cardiomyopathy. By integrating ECG, echo, MRI, CT, and omics data, it significantly outperforms standard clinical scores, boosting prediction accuracy (AUC increase up to 0.15). External validation across multiple cohorts confirmed generalizability and clinical usability. Proof of concept studies have been conducted integrating echocardiography strain imaging, MRI imaging of the epicardial adipose tissue and biological parameters to predict the risk of AF recurrency and for early detection of the atrial cardiomyopathy.
Conclusion
The MAESTRIA Demonstrator, built on a robust multimodal AI framework, promises to be a powerful clinical tool enabling clinicians to predict, visualize, and interpret cardiovascular risk, particularly atrial fibrillation, through the integration of clinical, imaging, and biological data.