Important Note: This site is designed to work best in Firefox, Safari or Chrome. Please update your browser to view the complete site.

Blog

Real Ways That Artificial Intelligence Will Change Cardiology

Posted on: 03.06.20

How Artificial Intelligence Will Change Cardiology

Artificial intelligence is developing rapidly in medicine, including in the field of cardiology.

If you’ve been to conferences in the last couple of years, you’ve probably seen presentations on some aspect of artificial intelligence. Quite simply, AI involves computer systems performing tasks that usually require human intelligence. They do this using techniques like deep machine learning and natural language processing.

AI is already seen in cardiology, performing tasks such as generating risk scores among patients. It is hoped that as AI develops further, it can help with improving the decision-making required on a daily basis, which has been proven to be subject to bias.

Here are some real ways that AI will change cardiology:

Counteract unconscious biases

Studies have proven that errors in decision-making can often come back to one of two types of bias. One of those is social bias, such as the categorizing of minorities or gender bias – in any case, this means that one group isn’t getting the same care as another.

In the case of cardiology, studies have shown that women tend to have a different experience than men, for example. Women are nearly twice as likely as men to die in the year following a heart attack, with a gendered difference in treatment being suspected in many cases.

The second type of bias that impacts errors in decision-making is “noise,” which means that decisions are impacted by irrelevant factors. For example, current mood, weather and the time since the last drink have all been found to impact decision-making.

The hope is that when AI is incorporated into daily decision-making in cardiology, it may help to counteract those unconscious biases. The caveat here is that the AI must be developed in such a way that it avoids bias. In some studies, AI has been found to amplify social bias in particular, especially in fields such as recruitment.

Get the bonus content: How Does AI Work?

One solution to this is better sampling. AI “learns” from the data it is given, so if that data is skewed toward people who fit within a certain category, then it can build-in the bias.

AI in cardiology decision-making is strictly a tool to aid the final decision, not to actually make it. The cardiologist should still retain the last step in the control of the system and have the authority to change the outcome recommended by the algorithm, especially in cases where the decision could go wrong.

Provide better risk prediction models

One exciting area undergoing development is the use of AI for patient risk prediction algorithms, where information is automatically extracted from medical imaging to offer better patient risk scoring.

Piotr J. Slomka, Ph.D., FACC, research scientist in the Artificial Intelligence in Medicine Program, Department of Medicine at Cedars-Sinai, and professor of medicine in-residence of the David Geffen School of Medicine, UCLA, is working with a Cedars-Sinai team to develop this.

From the Cedars-Sinai website: “Current research in the Slomka Laboratory focuses on developing innovative methods for fully automated analysis of nuclear cardiology data using novel algorithms and machine learning techniques, and on the development of integrated motion-corrected analysis of positron emission tomography (PET)/computed tomography (CT) angiography imaging.”

The image below illustrates how their proposed machine-learning model would allow prediction of the risk of major adverse cardiac events for a patient. This will be achieved by integrating clinical data with the imaging data.

Further information from the website states:

Specifically, the Slomka Laboratory aims to

-Develop new image-processing algorithms for fully automated analysis of fast-MPS. The algorithms will include better heart muscle detection by training with correlated anatomical data and a novel approach for mapping the probability of abnormal perfusion for each location of the heart muscle.

-Enhance the diagnosis of heart disease from fast-MPS by machine-learning algorithms that integrate clinical data, stress test parameters and quantitative image features.

-Demonstrate the clinical utility of the new algorithms applied to automatic canceling of the rest portion of the MPS scan when not needed.

ECGs and non-invasive glucose detection

One factor that can put patients off proper glucose monitoring is the finger prick test. AI is set to change this by detecting hypoglycemic events from raw ECG signals.

The technology involves a non-invasive, wearable sensor which works with the AI to detect blood glucose. “…using the latest findings of deep learning they can detect hypoglycemic events from raw ECG signals acquired with off-the-shelf non-invasive wearable sensors.”

The results of pilot studies found that sensitivity and specificity was around 82%, comparable to Continuous Glucose Monitoring devices.

Automated analysis of cardiac ultrasound

In November 2019, Ultromics received FDA clearance for their AI-drive echocardiography image analysis system, EchoGo Core.

“Traditionally, echocardiography has relied on the expert eye of clinicians, with years of experience, measuring the anatomical structures and identifying the disease, a potentially time-consuming and highly variable process. By automating the process and applying its AI analysis to look in greater detail at the scans, EchoGo enables clinicians to interpret echocardiograms efficiently and accurately and assists in their decision-making.

Get the bonus content: How Does AI Work?

EchoGo uses AI to calculate cardiac ultrasound left ventricular ejection fraction (EF), the most frequently used measurement of heart function, left ventricular volumes (LV) and, for the first time for an AI application, automated cardiac strain.” (Source)

Faster cardiac MRI analysis

A recent study revealed that cardiac MRI analysis can be performed significantly faster and with similar precision to a human expert when using automated machine learning.

When analyzing heart function on a cardiac MRI scan, it takes humans an average of 13 minutes, whereas with AI, the analysis takes around 4 seconds. Many decisions in cardiology rely on accurate and timely analysis of MRI, so if AI can do this accurately, it can be a game-changer.

The study wasn’t used to demonstrate the superiority of AI over humans and was not used to analyze patient outcomes from clinical assessment. However, it highlights great future potential for improved efficiencies.

The Impact of AI on Cardiology

Artificial intelligence (AI) being developed for healthcare has been one of the biggest topics at just about all medical conferences the past two years, and the FDA has already started the market clearance of several AI algorithms for medical imaging and other aspects of cardiology.

Importantly, it is expected that AI will help to improve outcomes in cardiology. It can help to speed up decision-making and counteract any unconscious biases. We’re betting we’ll see many more AI developments in the medical field in the coming years.



CT, MR and the Current State of Technology for Cardiac Imaging

Posted on: 08.08.19

Technology and the modalities that drive diagnostic cardiology continue to see innovation and development. New technologies and advancements in cardiac imaging will have a direct effect on both cardiologists and patients as they move mainstream. In this post, we’ll look at what is happening with cardiac imaging technology:

CT technology for Cardiac Imaging

One of the significant developments with CT Technology is CT with FRR. Fractional Flow Reserve (FRR) is traditionally an invasive, guide wire-based procedure allowing clinicians to accurately measure blood pressure and flow through a specific part of the artery. With CT, it is non-invasive.

A potential advantage of using FRR is that when combined with the CT, the images are powerful enough to identify whether a vessel needs medical intervention. Joel Sauer, Executive Vice President of Consulting at MedAxiom, has spoken with cardiologists who describe this technology as a “game-changer.”

Up until now, a decision to stent or not was generally based on the vessel being blocked above a certain percentage, however, this did not guarantee the procedure was the best course of action. For example, perhaps a vessel is clogged but isn’t the source of the issue for the patient, or a vessel might have build-up but is not worth stenting because muscle around it is already dead. FRR with CT can identify these situations and allow for a more effective treatment plan.

Get the bonus content: How AI Is Impacting Cardiology

Recent conference sessions on CT-FRR have discussed how the technology might now be mature enough to routinely use with patients. The Centers for Medicare and Medicaid Services (CMS) as well as private insurers have been approving reimbursements, driven by clinical evidence showing the technology can reduce the need for diagnostic angiograms or allow early discharge of patients who present in the emergency department (ED) where FFR-CT can definitively rule out severe ischemic heart disease.

The technology is not only useful for interventional planning, but for detecting and tracking disease early. This can mean that patients have the opportunity to reverse plaque formation and avoid surgery. The challenge will then be to keep them compliant for the years to come, another area in which CT has a role to play. Physicians can track patient condition using CT and use it to show where improvements have been made as a result of lifestyle changes, or where improvements could be made.

Challenges with Cardiac CT

Consequently, the technology has been slow to be adopted by medical practices. Part of this may be that the datasets have to be sent to a single provider, HeartFlow, via the internet. This is because they host the super-computing power required to run the powerful algorithms behind FFR-CT technology.

Another potentially limiting issue is that the scans are often read in the radiology department rather than a cardiologist. This requires extra training for radiologists for whom heart imaging may be a new field of study.

One thing that might spur further adoption is that use of FFR-CT can reduce the need for further testing. For example, where a patient has undergone stress testing and still has chest pain, a diagnostic cath would usually be ordered. FFR-CT can reduce the need for a cath where it won’t make any difference for the patient if it is used as the next step.

MRI technology for Cardiac Imaging

An MRI cardiac stress test is showing promising results, both for determining heart function and for predicting which cases are potentially fatal.

Cardiac magnetic resonance (CMR) has potential as a non-invasive alternative to tests such as catheterizations or nuclear imaging. In a study appearing in JAMA Cardiology, senior author Robert Judd, Ph.D., co-director of the Duke Cardiovascular Magnetic Resonance Center says:

“We’ve known for some time that CMR is effective at diagnosing coronary artery disease, but it’s still not commonly used and represents less than one percent of stress tests used in this country.

One of the impediments to broader use has been a lack of data on its predictive value — something competing technologies have,” Judd said. “Our study provides some clarity, although direct comparisons between CMR and other technologies would be definitive.”

Some barriers to further adoption of stress CMR include the availability of good-quality laboratories, a lack of data on outcomes and the necessity to exclude patients who cannot undergo magnetization. There’s also a similar issue to CT in that not everyone who can read nuclear imaging can read a CMR.

It is likely that many technologists will need special training to be able to produce the quality images that are needed. The study mentioned here provides a good starting point for head to head studies with other modalities, particularly to determine whether it is worth hospitals investing more in technology and training.

Get the bonus content: How AI Is Impacting Cardiology

Cardiac SPECT and PET

A clear advantage of nuclear SPECT is the availability of imagers and the overall cost of the procedure. When you compare SPECT to CT, it is more cost-effective and takes a lot less time. The fastest CT procedures (on top-of-the-line machines) generally take 45 minutes, whereas SPECT takes 30 minutes or less. Gamma technology has reduced imaging times for SPECT from 15-20 minutes to 2-4 minutes.

SPECT is utilized significantly more often than PET, particularly due to the high cost of the generator-based PET MPI agent and PET imaging systems..

There are a few noteworthy trends with SPECT and PET. In particular, there has been a marked decrease in numbers since the mid-2000s, despite most risk factors for cardiac events still being present. Secondly, there has been a decrease in abnormal test results over time.

There are a myriad of reasons indicated for these trends, such as the emergence of Appropriate Use Criteria in 2005 and more patients who can tolerate exercise undergoing stress testing without nuclear imaging as a first option.

In other developments, nonperfusion cardiac imaging is gaining ground. Here is an extract from a Journal of Nuclear Cardiology article, written by George A. Bellar:

The acceptance and growth of nonperfusion cardiac imaging applications are vital in widening the offerings of nuclear cardiology going forward. Already, much activity has been ongoing in validating the worth of F-18-flurodeoxyglucose (FDG) imaging for detecting and quantitating focal inflammatory lesions in patients with sarcoidosis.13,14

From 8 published studies, the pooled sensitivity and specificity values for detecting cardiac involvement with sarcoid were 89% and 78%, respectively. Combining MPI with FDG imaging could even enhance the accuracy of detection of sarcoid granulomas in the heart. Even combining PET with MRI for showing areas of delayed hyperenhancement in areas of decreased perfusion without inflammation could be an early manifestation of cardiac sarcoid.

Final thoughts

Bottom line – we’re seeing vast advances in the quality of imaging from all technologies and with that, the ability to accurately diagnose. However, each technology has strengths and weaknesses.

There’s the amount of time taken and, therefore, limitations placed on patient throughput. Then there are the relative costs and whether approval can be gained from CMS or insurance companies. In some cases pre-authorization is required.

Most of the time, it comes down to training and availability. Some of the more recent cardiac imaging technologies are not yet widely available or require additional training of technologists. Overall, it will be helpful to see more direct comparisons between the different modalities and their diagnostic qualities.

This post was written in collaboration with Joel Sauer, Executive Vice President of Consulting at MedAxiom.



Digirad — Revolutionary solid-state nuclear cardiology equipment and services.

Making Healthcare Convenient. As Needed. When Needed. Where Needed.