Feasibility and diagnostic accuracy of a deep learning algorithm for the evaluation of CADRADS classification with coronary computed tomography angiography

Congress:

ESCR 2019

Poster Number:

P-0073

Type:

Poster Abstract

Keywords:

Arteriosclerosis, CAD, CT, Cardiac

Authors:

G. Muscogiuri¹, M. Chiesa¹, M. Trotta¹, M. Guglielmo¹, A. Baggiano¹, A. I. Guaricci², M. Gatti³, S. Dell'aversana⁴, V. Palmisano⁵, G. rizzon⁶, A. Cavaliere⁷, A. Del Torto¹, P. Vivona¹, G. Cicala⁸, A. Loffreno¹, F. Ricci⁹, F. Baessato¹, L. fusini¹, G. Pontone¹; ¹Milan/IT, ²Bari/IT, ³Turin/IT, ⁴Naples/IT, ⁵Cagliari/IT, ⁶Padova/IT, ⁷Padua/IT, ⁸Parma/IT, ⁹Roma/IT

DOI:

DOI-Link:

https://dx.doi.org/

Fig. 2: Simulation of the diagnostic algorithm in a clinical setting using CNN for...

Fig. 1: A 53-year-old woman with family history of cardiovascular disease, palpitations...

Purpose

Artificial intelligence (AI) is increasing its role in diagnosis of patients with suspicious coronary artery disease. The aim of thisstudy is to develop a deep convolutional neural network (CNN) to classify coronary computed tomography angiography (CCTA) in the correct Coronary Artery Disease Reporting and Data System (CAD-RADS) category.

Methods and Materials

Two hundred eighty eight patients who underwent clinically indicated CCTA were included in this single-center retrospective study. The CCTAs were stratified byCAD-RADS scores by expert readers and considered as reference standard. A deep CNN was designed and tested on the CCTA dataset and compared to on-site reading. The deep CNN analyzed the diagnostic accuracy of the following three Models based on CADRADS classification: Model A (CADRADS 0 vs CADRADS 1-2 vs CADRADS 3,4,5), Model 1 (CADRADS 0 vs CADRADS>0), Model 2 (CADRADS 0-2 vs CADRADS...

Results

Model A showed a sensitivity, specificity, negative predictive value, positive predictive value and accuracy of 47%, 74%, 77%, 46% and 60%, respectively. Model 1 showed a sensitivity, specificity, negative predictive value, positive predictive value and accuracy of 66%, 91%, 92%, 63%, 86%, 89%, respectively. Conversely Model 2 demonstrated the following sensitivity, specificity, negative predictive value, positive predictive value and accuracy: 82%, 58%, 74%, 69%, 71%, 78%, respectively. Time of analysis was significantly lower using CNN as compared to on-site reading (530.5±179.1vs104.3±1.4 seconds, p:0.01)

Conclusion

Deep CNN yielded accurate automated classification of patients with CAD-RADS.