Deep learning for automatic calcium scoring in population based cardiovascular screening 

High volumes of standardized coronary artery calcium (CAC) scans are generated in screening that need to be scored accurately and efficiently to risk stratify individuals. Low-dose ECG triggered cardiac CT for calcium scoring of a population based screening study were used to externally validate the performance of deep-learning. In a screening setting, the reclassification rate was 2.0% for risk categorization 0-99, 100-399, ≥400 with very high agreement (κ=0.960) compared to manual assessment. Deep learning based software for automatic CAC scoring can be used in a cardiovascular CT screening setting with high accuracy for risk categorization and initiation of preventive treatment.

Fully automated deep learning powered calcium scoring in patients undergoing myocardial perfusion imaging

To assess the accuracy of fully automated deep learning (DL) based coronary artery calcium scoring (CACS) from non-contrast computed tomography (CT) as acquired for attenuation correction (AC) of cardiac single-photon-emission computed tomography myocardial perfusion imaging (SPECT-MPI).

In this study, 56 Patients who underwent cardiac SPECT-MPI due to suspected coronary artery disease (CAD) were prospectively enrolled. All patients underwent non-contrast CT for AC of SPECT-MPI twice. CACS was manually assessed (serving as standard of reference) on both CT datasets (n = 112) and by a cloud-based DL tool. The agreement in CAC scores and CAC score risk categories was quantified. Heart rate, image noise, body mass index (BMI), and scan did not significantly impact (p=0.09 – p=0.76) absolute percentage difference in CAC scores. Our study shows that a DL tool enables a fully automated and accurate estimation of CAC scores in patients undergoing SPECT-MPI. Thus, DL-based CACS may facilitate the further implementation of CAC scores as a routine imaging marker determined during the workup of SPECT-MPI examinations.