Current Status of Cardiac Imaging

Dr. Bosemani shared his perspectives on the current status of Cardiac Imaging in an email interview with us, which is reproduced below.

The advantage of coronary CT angiography is that it has the highest sensitivity of any modality to detect stenosis, with comparable specificity to alternatives. Coronary CTA can detect minimal or mild stenosis, which would show up as normal on a stress test.

Cardiac magnetic resonance (CMR) has emerged as new mainstream technique for the evaluation of patients with cardiac diseases, providing unique information to support clinical decision-making. Cardiovascular magnetic resonance (CMR), especially when employing late gadolinium enhancement (LGE) techniques, is a highly sensitive, non-invasive imaging modality for detecting alterations of the myocardium. CMR could be used to provide coronary artery imaging and reversible ischemia.

The coronary artery imaging by CMR is achievable with non-contrast whole-heart coronary magnetic resonance angiography (MRA) that can provide visualization of the coronary tree within a single three-dimensional acquisition with an average sensitivity, specificity and negative predictive value of 88%, 72% and 88%. Regarding the detection of reversible ischemia, the main advantage of CMR is the possibility to evaluate perfusion defects, wall motion abnormalities and viability without the use of ionizing radiation.

In this setting, CMR can be considered appropriate for diagnostic and prognostic purposes. Regarding prognostic stratification, the evidence of reversible perfusion defects on stress perfusion is the strongest independent predictor for cardiovascular events. Finally, stress CMR is appropriate in stable chest pain in patients with a previous history of revascularization and more cost-effectiveness as compared to an anatomical strategy.

Cardiac imaging has been growing faster than any other field of radiology in recent years. Mastering advanced cardiac imaging competencies require a combination of computed tomography (CT) and magnetic resonance (MR) physics knowledge, together with a deep understanding of cardiac anatomy, physiology and pathophysiology. As such, the acquisition of advanced skill sets is considered a longitudinal, if not a life-long, learning process, which is also strongly dependent on the availability of up-to-date scanners’ technology for clinical practice, the above factors make it less attractive for the residents and therefore resulting in shortage of cardiac radiologists.

Teleradiology can address the shortage by providing high quality reads by fellowship trained cardiac radiologists to both smaller radiology groups and larger institutions, where hiring subspecialty trained radiologists may not be a viable option. By providing cardiac teleradiology services, the organization may be able to expand non-invasive cardiology service which will benefit many patients.

AI-powered analysis of images and signals has reached human-level performance in many applications owing to the efficiency of modern machine learning methods, in particular deep learning using convolutional neural networks. Artificial intelligence (AI) algorithms have shown impressive results in specific and often time-consuming cardiovascular imaging tasks such as image segmentation, anomaly detection and patient selection.

CT fractional flow reserve (CT- FFR) is arising as a non-invasive alternative in diagnosing chest pain. Though in the early stages of clinical implementation, it is one of the few methods to provide an anatomical and functional assessment. ML algorithms can compute FFR without computational fluid dynamics and provide additional prognostic information. ML (Machine Learning) architectures have been used in a variety of different situations in CT such as analyzing cardiac chamber volume, LVH as well as coronary artery stenosis.

CMR is heralded as the gold standard for non-invasive assessment of the ejection fraction and left ventricular volume. It also enables tissue characterization which can dictate medical management. However, CMR takes substantial time in measuring volume or contour tracing. The integration of ML architectures can help expand the domain of CMR and transcend into new frontiers in cardiovascular imaging.