At this year's European Society of Cardiology’s 2022 Congress, clinicians are highlighting developments in cardiac imaging. What do we need to know about the latest advancements in cardiac imaging, and how may these advancements alter the field of cardiovascular diagnostics?
The European Society of Cardiology’s 2022 Congress took place in Barcelona from August 26th through the 29th, and this year’s event had a spotlight focus on cardiac imaging. So with that in mind, what do we need to know about the latest in cardiac imaging, and how may these advancements alter the field of cardiovascular diagnostics?
In the past decade, clinicians have seen developments in nuclear imaging, MRI, computed tomography (CT), and echocardiography. But let’s take a look at one of the latest developments: CT coronary angiography (CTCA).
The Clinical Utility of CT Coronary Angiography (CTCA)
This particular type of CT looks at the arteries that supply blood to the heart and the coronary arteries. Over the past decade, technological advances in this modality have improved sensitivity and negative predicative value for detecting significant coronary artery disease, and this demonstrated clinical utility has increased demand for this type of imaging technology.
The culmination of CTCA advancements led to its incorporation into the National Institute for Health and Care Excellence (NICE) guidelines for the assessment of recent onset chest pain. The guidelines have moved away from functional imaging and now recommends CTCA as the first-line approach for investigating the cause of both typical and atypical chest pain.
According to clinicians, these changes to the NICE guideline will help meet increasing demand for more imaging technology in clinical practice.
Machine Learning Applied to Clinical Imaging
Using computer-based algorithms to make decision, machine learning (ML) uses many variables without needing to understand the relationship of those variables to the outcome at the outset of the machine’s learning period. ML has many applications in the field of cardiology, but has become increasingly applied to clinical imaging.
Advances in this type of imaging have made it possible to analyze large data sets and extract relevant information, and with data sets from CTCA, ML can improve diagnoses and help better predict functionally significant lesions. On top of this, advancements in cardiac imaging will be able to automatically quantify markers, such as calcium scoring and epicardial fat volume, and add this data into scoring systems.
The implementation of machine learning in clinical practice can help cardiologists improve risk prediction. In a feasibility study, clinicians analyzed 10,030 patients from a large study with five-year follow-up data and used 25 clinical and 44 CTCA parameters for predicting risk. The results of the study showed that ML was better at predicant mortality than any individual clinical or CTCA-based risk factor.
Now while the study didn’t account for more sophisticated risk stratification calculators such as Astro-CHARM or MESA, this data demonstrated the clinical utility of machine learning in the cardiology field.
Echocardiology as a Diagnostic Tool
More than 60 years have passed since clinicians began using echocardiology as a diagnostic tool for cardiovascular disease. But since then, this tool has evolved from a simple M-mode imaging technique to two-dimensional and three-dimensional technology, pulsed and continuous wave doppler, color flow and tissue Doppler, and transesophageal echocardiography.
But one of the latest developments is adaptive contrast enhancement (ACE), which is a software-based beamforming technique that stores multiple sequential datasets for each probe element before analyzing it. Using this software allows clinicians to strengthen the image pixel from the real structure and suppresses the image pixel from noise or artifact to obtain a high-resolution image.
Researchers at the Mount Sinai Icahn School of Medicine in New York assessed the use of echocardiographic images obtained with ACE by comparing them to images obtained using standard hardware-based beamforming techniques. After performing this assessment, they found that the software-based beamformer technique with the ACE algorithm significantly improved visualization of endocardial borders in the following:
- Anteroseptal wall segment
- Anterolateral wall segment
- Inferolateral wall segment
- Inferior wall segment
- Anterior wall segment
They also found that the ACE technology can reduce medical costs by decreasing the need for contrast usage and reduced the need for additional diagnostic testing.
As developments in cardiac imaging continue to emerge, clinicians can adopt these tools into practice to improve the diagnosis and treatment of cardiovascular disease for patients.
References:
Thomas D Heseltine, Scott W Murray, Balazs Ruzsics, Michael Fisher, Latest Advances in Cardiac CT, European Cardiology Review 2020;15:e01. https://doi.org/10.15420/ecr.2019.14.2
Wang CL, Hung KC. Recent Advances in Echocardiography. J Med Ultrasound. 2017 Apr-Jun;25(2):65-67. doi: 10.1016/j.jmu.2017.03.010. Epub 2017 May 9. PMID: 30065461; PMCID: PMC6029319.