We Got the Beat: 5 Ways AI Can Help Clinicians Care for CVD Patients

Cardiovascular disease (CVD), which refers to any disease of the heart, vascular disease of the brain, or disease of the blood vessel, is the leading cause of death and disability in the world today.

While more people die from CVDs world wide than from any other cause, the approximately 485.6 million[i] people living with CVD face the risk of further health complications introduced by additional disease, environmental conditions, and now COVID-19.

Even in a non-COVID world, each cardiology patient presents a unique dataset for physicians to sort through as diagnostic imaging, electronic health records, and remote monitoring devices become mainstream.[ii],[iii] Acquiring and integrating each patient’s vast amount of data into coherent, actionable, and personalized treatment in real-time is reaching an unmanageable level of complexity.ii But as data analysis becomes increasingly sophisticated, one strategy that physicians are using to effectively combat the overload of health-related data is artificial intelligence (AI).[iv]

With new viruses like COVID-19 affecting and exacerbating CVD patients’ existing conditions, the healthcare community also finds itself challenged to understand and manage this new dilemma. Some emerging guidelines include:

  • Patients need to be triaged based on underlying health risks as part of the process to identify patients with a higher likelihood of developing a severe form of COVID-19 so targeted care can be implemented.
  • Special attention must be given to ensuring that there are separate facilities in place for dealing with COVID-19 cardiac patients and non-COVID-19 cardiac patients including catheterization laboratories for performing invasive heart examinations.

Now, more than ever, advances in AI-enabled cardiac imaging technology are needed to assist the medical community with managing the way patients are triaged, diagnosed, treated, and monitored. Here are five ways AI-enabled technologies are being used to help cardiac patients:

1. An electrocardiogram (ECG) records the electrical activity of the heart at rest. It provides information about heart rate and rhythm and shows if there is enlargement of the heart due to high blood pressure (hypertension) or evidence of a previous heart attack (myocardial infarction). Acquisition of quality ECG data is critical in providing accurate and timely diagnosis and patient treatment, and a significant challenge that hospitals face is duplicate ECGs. On GE Healthcare’s MAC VU360, Smart Lead technology automatically detects a new patient to help minimize patient data mix-ups. An enhanced hook-up advisor guides new users to a clean, high-quality waveform. Smart Auto-ECG automatically captures and displays the first clean, high quality waveform, reducing repeats.

conventional cardiac.jpgCardiac with air recon.jpg

Conventional Cardiac MRI

Cardiac with AIR Recon DL

2. Cardiac magnetic resonance imaging (MRI) is commonly used when a comprehensive assessment of the heart is needed due to its versatility. MRI is a powerful tool to measure blood flow, visualize tissue pathology and thus specifically verifying or refuting the underlaying cause of cardiac disease. However, MRI historically has an inherent compromise between image quality and scan time. Better image quality achieved through higher signal-to-noise (SNR) and/or spatial resolution needed to show anatomical detail necessitated long scan times. Shorter scans, aimed to improve patient comfort and productivity, compromised image quality and diagnostic confidence.

With new MRI technology, AIR Recon DL, clinicians and technologists will no longer have to choose between image quality and scan time. In addition to improving SNR, this technology features a unique intelligent ringing suppression that preserves fine image details, helping address two common pain points for radiologists and technologists—image noise and ringing. AIR Recon DL is a deep learning-based reconstruction engine that makes full use of raw data for maximum image quality. This new technology can help with several challenging cardiac MR sequences used to assess CVD by reducing scan time and improving image quality.

AI Auto Measure 2D

AI Auto Measure 2D

3. Echocardiography is the standard in diagnosis, management and follow-up of patients with heart disease and assessing the interactions between COVID-19 and the heart. However, it requires methodical and time-consuming assessments of heart function. Furthermore, patient-related factors such as size and echogenicity may impact image quality and measurement accuracy.[v] Therefore, high quality data acquisition and operator skill are key elements to achieve accurate and complete exams. And, as patients undergo subsequent monitoring exams, the reproducibility of exam assessments is key to identifying improvement or disease progression.

GE Healthcare’s new cardiovascular ultrasound package, Vivid Ultra Edition leverages AI-driven, neural network-based algorithms designed to deliver repeatable and faster measurements in 2D echo imaging. AI Auto Measure – Spectrum Recognition can semi-automatically detect the appropriate measurement of spectral Doppler images, enabling the system to fast-forward the path from scanning to measurements. AI Auto Measure – 2D can detect the relevant points in the image used to derive key measurements of the heart’s left ventricle, performing comparable to human users with 100% reproducibility.vii Exam time is reduced, with up to 80% less clicks to get 2D measurements, and inter-operator variability diminished.[vi]


True Fidelity CT

4. Earlier this month, the American College of Cardiology (ACC) published a report noting that Coronary CT Angiography (CCTA) can be used as a first-line evaluation for coronary artery disease (CAD), a type of cardiovascular disease. Traditionally, CAD patients are assessed using invasive cine-angiography (commonly referred to as an angiogram) but the ACC’s report now supports CCTA as a less invasive option. Using this form of CT imaging, clinicians can examine the arteries that supply blood to the heart, helping clinicians diagnose the cause of chest pain and other symptoms such as plaque buildup that may stop blood flow and cause a heart attack. The test uses an advanced 3D X-ray machine to produce images of the heart and its blood vessels.

Deep Learning Image Reconstruction helps elevate these images using a dedicated Deep Neural Network to generate TrueFidelity CT Images[vii], which have the potential to improve the reading confidence in cardiovascular and many other clinical applications. Compared to current iterative reconstruction technology, TrueFidelity CT Images can elevate every image to a powerful first impression with impressive image quality performance[1], and preferred image sharpness and noise texture[2], without compromising dose performance.

5. While point of care ultrasound has been shown to provide many benefits for bedside assessment, a quick view with handheld ultrasound at the point of care is still beneficial to assess the patient’s condition quickly so they can get to the right place at the right time. AI-based software can help automate the analysis process. For example, LVivo EF on GE Healthcare’s Vscan Extend provides automated and objective information for evaluating Ejection Fraction, which enables clinicians with various levels of ultrasound experience and training to work more effectively and consistently. Click for more on how AI-enhanced cardiac analysis with Vscan Extend ultrasound is empowering clinicians on the frontlines.

AI-powered medical technology is already proving helpful to improving cardiac care. However, there is further opportunity to seamlessly integrate such technologies to help heart care teams make sound diagnostic decisions and operate at peak efficiency while delivering high-quality patient care.

[1] Image quality comparisons between DLIR and ASiR-V, were evaluated by phantom tests of MTF, SSP, axial NPS, standard deviation of image noise, CT Number accuracy, CNR, and artefact analysis.  Additionally, LCD was demonstrated in phantom testing using a model observer with the head and body MITA CT IQ Phantoms (CT191, CT189 The Phantom Laboratory). DLIR and ASiR‐V reconstructions were performed using the same raw data.

[2] As demonstrated in a clinical evaluation consisting of 60 cases and 9 physicians, where each case was reconstructed with both DLIR and ASiR‐V and evaluated by 3 of the physicians. In 100% of the reads, DLIR’s image sharpness was rated the same as or better than ASiR‐V’s.  In 91% of the reads, DLIR’s noise texture was rated better than ASiR‐V’s.  This rating was based on each individual reader’s preference.

[i] Heart Disease and Stroke Statistics—2020 Update: A Report From the American Heart Association. Originally published 29 Jan 2020Circulation. 2020;141:e139–e596.

[ii] Outlining How Artificial Intelligence May Help Adhere to Cardiac Care Guidelines. Diagnostic and Interventional Cardiology Accessed 12/21/2018

[iii] Lost in Thought — The Limits of the Human Mind and the Future of Medicine. NEJM New England Journal of Medicine. Accessed 12/21/2018

[iv] Artificial Intelligence in Cardiology. JACC Journal of the American College of Cardiology Accessed 12/21/2018

[v] Temporal Trends in the Utilization of Echocardiography in Ontario, 2001 to 2009, Blecker et al, JACC: Cardiovascular Imaging Volume 6, Issue 4, April 2013, Pages 515-522,

[vi] The Role of AI in Streamlining Echocardiography Quantification White Paper, Kristin McLeod – JB80498XX

[vii] TrueFidelity CT Images are available on GE Healthcare’s Revolution Apex and Revolution EVO Gen 3 CT systems as well as an upgrade on the company’s Revolution CT scanner