Detection of coronary stenosis is of utmost importance in identifying vulnerable patients. The combined use of coronary computed tomography angiography at rest (CCTA) and stress myocardial computed tomography perfusion (stress-CTP) provides both anatomic and functional analysis of coronary artery disease (CAD) using a single imaging test. Stress-CTP evaluates myocardial perfusion by measuring myocardial blood flow (MBF) under pharmacologically induced stress conditions. The drawback is that stress-CTP requires additional scanning and administration of an intravenous stressor with an increase in radiation exposure and potential stressor-related side effects. The investigators recently patented a computational model that can reproduce MBF under stress conditions (Italian patent n. 102021000031475 Metodo implementato mediante computer per la simulazione del flusso sanguigno miocardico in condizioni di stress \[Computational method for simulating myocardial blood flow in stress conditions\], half owned by Centro Cardiologico Monzino, half by Politecnico di Milano).
On top of this, CCTA can characterize plaque type and identify adverse plaque characteristics. Moreover, biomechanics analysis allows the study of luminal stenosis and stress within the plaque. Finally, radiomics, extracting quantitative features from medical images to create big data and identify novel imaging biomarkers, can be applied to improve the diagnostic accuracy of coronary plaques.
Symptomatic patients with suspected CAD referred for nonemergent, clinically indicated non-invasive coronary angiography.
Low pre-test likelihood of CAD
Prior myocardial infarction
Previous history of revascularization
Acute coronary syndrome
A total of 400 consecutive symptomatic patients with suspected CAD who were recruited for non-emergent, clinically indicated non-invasive coronary angiography with eventually dynamic stress computed tomography perfusion study
Cohort
Other
Not listed
Development of a computational model to predict MBF avoiding CT stress protocol
CONCERTO main aim is to predict MBF values from cardiac CT scans without stress protocol. This aim will be achieved by improving an existing computational model. To achieve this goal, we first need to explore available information on the effect of stenosis on the pressure gradient across it and the division of flow in the coronary tree. A second step will be a ML analysis on a reasonable number of patients with known myocardial perfusion (from CT-stress) to build a suitable neural network that can predict some general features of the model parameters. This neural network will allow us to include this information in our calibration procedure and use also it for future patients in whom stress CT will no longer be necessary. Finally, a comprehensive, patient-specific model calibration strategy will be developed leveraging the results. This will allow us to apply our tool to any patient as long as a standard CT acquisition and some pressure
Improvement of CAD risk assesment
The second Aim is to improve CAD risk assessment by integrating the information from the perfusion computation model with imaging and radiomics features
Click on the lead sponsor to view all of their trials.
Centro Cardiologico Monzino
Lead sponsor
Politecnico di Milano
Collaborator
University of Bari
Collaborator
