Giuseppe Muscogiuri, Mattia Chiesa, Andrea Baggiano, Pierino Spadafora, Rossella De Santis, Marco Guglielmo, Stefano Scafuri, Laura Fusini, Saima Mushtaq, Edoardo Conte, Andrea Annoni, Alberto Formenti, Maria Elisabetta Mancini, Francesca Ricci, Francesco Paolo Ariano, Luigi Spiritigliozzi, Mario Babbaro, Rocco Mollace, Riccardo Maragna, Carlo Maria Giacari, Daniele Andreini, Andrea Igoren Guaricci, Gualtiero I. Colombo, Mark G. Rabbat, Mauro Pepi, Francesco Sardanelli, Gianluca Pontone, Muscogiuri, G, Chiesa, M, Baggiano, A, Spadafora, P, De Santis, R, Guglielmo, M, Scafuri, S, Fusini, L, Mushtaq, S, Conte, E, Annoni, A, Formenti, A, Mancini, M, Ricci, F, Ariano, F, Spiritigliozzi, L, Babbaro, M, Mollace, R, Maragna, R, Giacari, C, Andreini, D, Guaricci, A, Colombo, G, Rabbat, M, Pepi, M, Sardanelli, F, and Pontone, G
Purpose: Artificial intelligence could play a key role in cardiac imaging analysis. To evaluate the diagnostic accuracy of a deep learning (DL) algorithm predicting hemodynamically significant coronary artery disease (CAD) by using a rest dataset of myocardial computed tomography perfusion (CTP) as compared to invasive evaluation. Methods: One hundred and twelve consecutive symptomatic patients scheduled for clinically indicated invasive coronary angiography (ICA) underwent CCTA plus static stress CTP and ICA with invasive fractional flow reserve (FFR) for stenoses ranging between 30% and 80%. Subsequently, a DL algorithm for the prediction of significant CAD by using the rest dataset (CTP-DLrest) and stress dataset (CTP-DLstress) was developed. The diagnostic accuracy for identification of significant CAD using CCTA, CCTA+CTPStress, CCTA+CTP-DLrest, and CCTA+CTP-DLstress were measured and compared. The time of analysis for CTPStress, CTP-DLrest and CTP-DLStress were recorded. Results: Patient-specific sensitivity, specificity, NPV, PPV, accuracy and area under the curve (AUC) of CCTA alone and CCTA+CTPStress were 100%, 33%, 100%, 54%, 63%, 67% and 86%, 89%, 89%, 86%, 88%, 87%, respectively. Patient-specific sensitivity, specificity, NPV, PPV, accuracy and AUC of CCTA+DLrest and CCTA+DLstress were 100%, 72%, 100%, 74%, 84%, 96% and 93%, 83%, 94%, 81%,88%,98%, respectively. All CCTA+CTPStress, CCTA+CTP-DLRest and CCTA+CTP-DLStress significantly improved detection of hemodynamically significant CAD (pConclusion: Evaluation of myocardial ischemia using a DL approach on rest CTP datasets is feasible and accurate. This approach may be a useful gatekeeper prior to CTPStress.