Maximization of the usage of coronary CTA derived plaque information using a machine learning based algorithm to improve risk stratification; insights from the CONFIRM registry


Alexander R. van Rosendael, New York Presbyterian Hospital
Gabriel Maliakal, New York Presbyterian Hospital
Kranthi K. Kolli, New York Presbyterian Hospital
Ashley Beecy, New York Presbyterian Hospital
Subhi J. Al'Aref, New York Presbyterian Hospital
Aeshita Dwivedi, New York Presbyterian Hospital
Gurpreet Singh, New York Presbyterian Hospital
Mohit Panday, New York Presbyterian Hospital
Amit Kumar, New York Presbyterian Hospital
Xiaoyue Ma, New York Presbyterian Hospital
Stephan Achenbach, Friedrich-Alexander-Universität Erlangen-Nürnberg
Mouaz H. Al-Mallah, King Saud bin Abdulaziz University for Health Sciences
Daniele Andreini, IRCCS Centro Cardiologico Monzino
Jeroen J. Bax, Leiden University Medical Center - LUMC
Daniel S. Berman, Cedars-Sinai Medical Center
Matthew J. Budoff, The Lundquist Institute
Filippo Cademartiri, IRCCS Fondazione SDN
Tracy Q. Callister, Tennessee Heart and Vascular Institute
Hyuk Jae Chang, Yonsei University College of Medicine
Kavitha Chinnaiyan, William Beaumont Hospital
Benjamin J.W. Chow, University of Ottawa, Canada
Ricardo C. Cury, Baptist Cardiac and Vascular Institute
Augustin DeLago, Capitol Cardiology Associates
Gudrun Feuchtner, Medizinische Universitat Innsbruck
Martin Hadamitzky, Deutsches Herzzentrum München
Joerg Hausleiter, Ludwig-Maximilians-Universität München
Philipp A. Kaufmann, UniversitatsSpital Zurich
Yong Jin Kim, Seoul National University Hospital
Jonathon A. Leipsic, The University of British Columbia
Erica Maffei
Hugo Marques, Hospital da Luz
Gianluca Pontone, IRCCS Centro Cardiologico Monzino
Gilbert L. Raff, William Beaumont Hospital

Document Type


Publication Date


Publication Title

Journal of Cardiovascular Computed Tomography


© 2018 Introduction: Machine learning (ML) is a field in computer science that demonstrated to effectively integrate clinical and imaging data for the creation of prognostic scores. The current study investigated whether a ML score, incorporating only the 16 segment coronary tree information derived from coronary computed tomography angiography (CCTA), provides enhanced risk stratification compared with current CCTA based risk scores. Methods: From the multi-center CONFIRM registry, patients were included with complete CCTA risk score information and ≥3 year follow-up for myocardial infarction and death (primary endpoint). Patients with prior coronary artery disease were excluded. Conventional CCTA risk scores (conventional CCTA approach, segment involvement score, duke prognostic index, segment stenosis score, and the Leaman risk score) and a score created using ML were compared for the area under the receiver operating characteristic curve (AUC). Only 16 segment based coronary stenosis (0%, 1–24%, 25–49%, 50–69%, 70–99% and 100%) and composition (calcified, mixed and non-calcified plaque) were provided to the ML model. A boosted ensemble algorithm (extreme gradient boosting; XGBoost) was used and the entire data was randomly split into a training set (80%) and testing set (20%). First, tuned hyperparameters were used to generate a trained model from the training data set (80% of data). Second, the performance of this trained model was independently tested on the unseen test set (20% of data). Results: In total, 8844 patients (mean age 58.0 ± 11.5 years, 57.7% male) were included. During a mean follow-up time of 4.6 ± 1.5 years, 609 events occurred (6.9%). No CAD was observed in 48.7% (3.5% event), non-obstructive CAD in 31.8% (6.8% event), and obstructive CAD in 19.5% (15.6% event). Discrimination of events as expressed by AUC was significantly better for the ML based approach (0.771) vs the other scores (ranging from 0.685 to 0.701), P < 0.001. Net reclassification improvement analysis showed that the improved risk stratification was the result of down-classification of risk among patients that did not experience events (non-events). Conclusion: A risk score created by a ML based algorithm, that utilizes standard 16 coronary segment stenosis and composition information derived from detailed CCTA reading, has greater prognostic accuracy than current CCTA integrated risk scores. These findings indicate that a ML based algorithm can improve the integration of CCTA derived plaque information to improve risk stratification.





First Page


Last Page






This document is currently not available here.