Artificial Intelligence in Cardiology

 

We are doing active research in applying Machine Learning (ML) in Medicine. In particular we are really proud of a large research project in which we were involved for the ML part including model development, model comparison and result analysis. This collaboration project has resulted in an important paper published in The Lancet in January 2021 (https://doi.org/10.1016/S0140-6736(20)32519-8). Its reference is 

F.D’Ascenzo, D. O. Filippo, G. Gallone, G. Mittone, M. A. Deriu, M. Iannaccone, A. Ariza-Solé, C. Liebetrau, S. Manzano-Fernández, G. Quadri, T. Kinnaird, G. Campo, J. S. Henriques, J. Hughes, A. Dominguez-Rodriguez, M. Aldinucci, U. Morbiducci, G. Patti, S. Raposeiras-Roubin, E. Abu-Assi, D. G. M. Ferrari, F. Piroli, A. Saglietto, F. Conrotto, P. Omedé, A. Montefusco, M. Pennone, F. Bruno, P. P. Bocchino, G. Boccuzzi, E. Cerrato, F., O. Varbella, M. Sperti, S. B. Wilton, L. Velicki, I. Xanthopoulou, A. Cequier, A. Iniguez-Romo, M. I. Pousa, C. M. Fern, C. B. Queija, R. Cobas-Paz, A. Lopez-Cuenca, A. Garay, F. P. Blanco, A. Rognoni, B. G. Zoccai, S. Biscaglia, I. Nunez-Gil, T. Fujii, A., ro Durante, X. Song, T. Kawaji, D. Alexopoulos, Z. Huczek, J. R. G. Juanatey, S. -P. Nie, M. Kawashiri, I. Colonnelli, B. Cantalupo, R. Esposito, S. Leonardi, W. G. Marra, A. Chieffo, U. Michelucci, D. Piga, M. Malavolta, S. Gili, M. Mennuni, C. Montalto, L. O. Visconti, and Y. Arfat, “Machine learning-based prediction of adverse events following an acute coronary syndrome (praise): a modelling study of pooled datasets,” The Lancet, vol. 397, iss. 10270, p. 199–207, 2021.

Or if you prefer the BibTeX entry you can find it at the end of the page.  

Score web Calculator

If you are interested in trying the score developed you can check it out at the URL: https://praise.hpc4ai.i

 

Summary from the paper of the research work

 

It was a large study in which we developed a machine learning-based risk stratification model to predict all-cause death, recurrent acute myocardial infarction, and major bleeding after ACS.

 Different machine learning models for the prediction of 1-year post-discharge all-cause death, myocardial infarction, and major bleeding (defined as Bleeding Academic Research Consortium type 3 or 5) were trained on a cohort of 19826 adult patients with ACS (split into a training cohort [80%] and internal validation cohort [20%]) from the BleeMACS and RENAMI registries, which included patients across several continents. 25 clinical features routinely assessed at discharge were used to inform the models. The best-performing model for each study outcome (the PRAISE score) was tested in an external validation cohort of 3444 patients with ACS pooled from a randomised controlled trial and three prospective registries. Model performance was assessed according to a range of learning metrics including area under the receiver operating characteristic curve (AUC).

The PRAISE score showed an AUC of 0·82 (95% CI 0·78–0·85) in the internal validation cohort and 0·92 (0·90–0·93) in the external validation cohort for 1-year all-cause death; an AUC of 0·74 (0·70–0·78) in the internal validation cohort and 0·81 (0·76–0·85) in the external validation cohort for 1-year myocardial infarction; and an AUC of 0·70 (0·66–0·75) in the internal validation cohort and 0·86 (0·82–0·89) in the external validation cohort for 1-year major bleeding.

A machine learning-based approach for the identification of predictors of events after an ACS is feasible and effective. The PRAISE score showed accurate discriminative capabilities for the prediction of all-cause death, myocardial infarction, and major bleeding, and might be useful to guide clinical decision making.

Feature Importance Plots. Figure taken from the paper.

The models studied obtained a very promising performance. More details in the paper.

BibTex Entry

@article{2021machine,
title={Machine learning-based prediction of adverse events following an acute coronary syndrome (PRAISE): a modelling study of pooled datasets},
author={F.D'Ascenzo and O. De Filippo and G. Gallone and G. Mittone and M.A. Deriu and M. Iannaccone and A. Ariza-Solé and C. Liebetrau and S. Manzano-Fernández and G. Quadri and T. Kinnaird and G. Campo and J. S. Henriques and J. Hughes and A. Dominguez-Rodriguez and M. Aldinucci and U. Morbiducci and G. Patti and S. Raposeiras-Roubin and E. Abu-Assi and G.M. De Ferrari and F. Piroli and A. Saglietto and F. Conrotto and P. Omedé and A. Montefusco and M. Pennone and F. Bruno and P. Paolo Bocchino and G. Boccuzzi and E. Cerrato and F. and O. Varbella and M. Sperti and S. B. Wilton and L. Velicki and I. Xanthopoulou and A. Cequier and A. Iniguez-Romo and I. Munoz Pousa and M. Cespon Fern and B. Caneiro Queija and R. Cobas-Paz and A. Lopez-Cuenca and A. Garay and P. Flores Blanco and A. Rognoni and G. Biondi Zoccai and S. Biscaglia and I. Nunez-Gil and T. Fujii and A. and ro Durante and X. Song and T. Kawaji and D. Alexopoulos and Z. Huczek and J. R. G. Juanatey and S.-P. Nie and M.-aki 
Kawashiri and I. Colonnelli and B. Cantalupo and R. Esposito and S. Leonardi and W. G. Marra and A. Chieffo and U. Michelucci and D. Piga and M. Malavolta and S. Gili and M. Mennuni and C. Montalto and L. O. Visconti and Y. Arfat}, journal={The Lancet}, volume={397}, number={10270}, pages={199--207}, year={2021}, url={https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(20)32519-8/fulltext} }