Journal Article

A machine-learning based bio-psycho-social model for the prediction of non-obstructive and obstructive coronary artery disease

Sperduti, Nicolò, Romagnoli, Tommaso, De Giorgi, Alfredo, Visioli, Giacomo, Polastri, Michele, Gioffrè, Laura, Gloria, Brombo, Quattrino, Silvia, Moretti, Andrea, Miraldi, Fabio, Consorti, Fabrizio, Kaur, Amanpreet, Cutini, Andrea, Zuliani, Giovanni, Salis, Paolo, Toriello, Filippo, Amoroso, Daria, Romiti, Giulio Francesco, Scotti, Biagio, Cocco, Nino, Rossoni, Agostino, Flego, Davide, Rosa, Antonello, Borgi, Marco, Isidori, Andrea, Zito, Andrea, Tomassi, Marta, Proietti, Marco, Cantelmi, Claudio, Rozzi, Gloria, Tanzilli, Alessandra, Manfredini, Roberto, Ceccarelli, Simona, Petriello, Gennaro, Cangemi, Roberto, Mazzonna, Piergiovanni, Scacciavillani, Roberto, Barillà, Francesco, tunato, Valeria, Servello, Adriana, Galardo, Gioacchino, Tellini, Laurent, Nocchi, Silvia, Santangelo, Giuseppe, Palladino, Mariangela, Vincenti, Marianna, Costabile, Gianluca, Morelli, Sergio, Dalla Nora, Edoardo, Marrapodi, Ramona, Tiberti, Claudio, Corica, Bernadette, Passaro, Angelina, Di Iorio, Martina, Spaggiari, Riccardo, Alessandroni, Maria, Di Vincenzo, Alessandra, Vellucci, Ilaria, Segatori, Antonio, Testa, Floriana, M

2023

Overview

BackgroundMechanisms of myocardial ischemia in obstructive and non-obstructive coronary artery disease (CAD), and the interplay between clinical, functional, biological and psycho-social features, are still far to be fully elucidated. ObjectivesTo develop a machine-learning (ML) model for the supervised prediction of obstructive versus non-obstructive CAD.MethodsFrom the EVA study, we analysed adults hospitalized for IHD undergoing conventional coronary angiography (CCA). Non-obstructive CAD was defined by a stenosis < 50% in one or more vessels. Baseline clinical and psycho-socio-cultural characteristics were used for computing a Rockwood and Mitnitski frailty index, and a gender score according to GENESIS-PRAXY methodology. Serum concentration of inflammatory cytokines was measured with a multiplex flow cytometry assay. Through an XGBoost classifier combined with an explainable artificial intelligence tool (SHAP), we identified the most influential features in discriminating obstructive versus non-obstructive CAD. ResultsAmong the overall EVA cohort (n = 509), 311 individuals (mean age 67 ± 11 years, 38% females; 67% obstructive CAD) with complete data were analysed. The ML-based model (83% accuracy and 87% precision) showed that while obstructive CAD was associated with higher frailty index, older age and a cytokine signature characterized by IL-1β, IL-12p70 and IL-33, non-obstructive CAD was associated with a higher gender score (i.e., social characteristics traditionally ascribed to women) and with a cytokine signature characterized by IL-18, IL-8, IL-23.ConclusionsIntegrating clinical, biological, and psycho-social features, we have optimized a sex- and gender-unbiased model that discriminates obstructive and non-obstructive CAD. Further mechanistic studies will shed light on the biological plausibility of these associations.Clinical trial registrationNCT02737982.

• Publisher: Springer Nature B.V
• Subject: Angiography / Artificial intelligence / Cardiovascular disease / Coronary artery disease / Coronary vessels / Cultural characteristics / Cytokines / Explainable artificial intelligence / Flow cytometry / Gender / Heart diseases / Inflammation / Interleukin 12 / Interleukin 18 / Interleukin 23 / Interleukin 8 / Ischemia / Learning algorithms / Machine learning / Model accuracy / Myocardial ischemia / Social discrimination learning / Stenosis / Vein & artery diseases