Joshua Farkas, MD regarding recent reports of MI’s troponin-driven overdiagnosis issue.

Artificial Intelligence

Cedars-Sinai’s Massive Echo + Reporting Foundation Model

The team at Cedars-Sinai’s Smidt Heart Institute has become among the most prolific health system-based cardiovascular AI developers, and they added to their AI resume after unveiling the massive new EchoCLIP foundation model, which combines echo images and reporting text to perform a wide range of interpretations — without specific training.

• Foundation models are a type of generative AI, using a vast amount of unlabeled data to perform a wider range of clinical tasks than we’ve seen with most current healthcare AI tools.

To develop this beast of an echo AI model, the team assembled a dataset of 1,032,975 cardiac ultrasound videos and corresponding text-based expert interpretations.

Even without task-specific training or fine-tuning, EchoCLIP performed well across a wide range of measurement and detection tasks when tested against external data, including:

• Accessing cardiac function by predicting LVEF (7.1% mean absolute error)
• Estimating pulmonary artery pressure (10.8 mm Hg mean absolute error)
• Identifying intracardiac devices, like mitral valve repair, TAVR, and pacemaker/defibrillator leads (AUCs = 0.97, 0.92, 0.84)
• Detecting changes from a healthy cardiac chamber size, like severe dilation of the right ventricle, right atrium, left ventricle, and left atrium (AUCs = 0.92, 0.97, 0.92, 0.91)
• Assessing tamponade and severe left ventricular hypertrophy (AUCs = 0.96 & 0.82)

Perhaps more impressively, the team’s related EchoCLIP-R system was able to accurately identify specific patients using only their exams and retrieve past exams (AUC = 0.86), while highlighting clinically important changes that occurred between their echos.

Altogether, these results suggest that with a large enough dataset of echo images and expert text interpretations we can train foundational models that can support an extremely wide range of echo assessment tasks. 

The Takeaway

The last few years have brought an impressive flow of echo AI models, and Cedars-Sinai’s new EchoCLIP model certainly could prove to be among the most significant given its size, breadth of capabilities, and its role as the first of potentially many advanced echo image+text foundation models.

Accurate and Efficient Remote Cardiac Patient Monitoring

The expansion of remote cardiac patient monitoring is creating more care opportunities, but also new operational challenges for cardiology teams. Check out this Cardiac Wire Show, where ARTELLA Solutions’ Jacinta Fitzsimons shares how the right combination of technology and service can help physicians get the most out of their cardiac RPM programs – today and into the future.

It’s Time to See The Full Picture

See how HeartFlow ONE is transforming precision heart care as the first all-in-one CCTA pathway, combining FFRCT, stenosis, and plaque analysis in a single workflow.

Merge and Duly Health Streamline Cardiology Reporting

Over the last 10 years, Dr. Sujith Kalathiveetil of Duly Health and Care has seen a significant evolution in cardiovascular imaging and experienced a similar evolution with Merge’s cardiology solutions. See how Merge Cardio has helped make cardiology reporting more consistent, accurate, and easier to obtain for Dr. Kalathiveetil and his colleagues.

The Wire

• Beyond the Plaque: New research suggests that atherosclerosis shares similarities with cancer, as smooth muscle cells in atherosclerosis display tumor-like traits like hyperproliferation and resistance to cell death. Interestingly, targeting DNA repair pathways with an anti-cancer drug, niraparib, showed promise in slowing atherosclerosis progression. Research in atherosclerosis is abundant, but new ways of understanding the disease continue to emerge and may reshape our approach to combating it.

• Women Undertreated: A registry study found that Swedish women with heart disease are less likely to get treated with cholesterol-lowering drugs for primary prevention than their male counterparts. In the study of electronic health records from 1,452 people diagnosed with chronic coronary artery disease between 2012 and 2020, 54% of women were treated compared with 74% of men. Women were also more likely to discontinue taking their medication within three years and more likely to miss cholesterol targets. 

• ECG AI Saves Lives: A new study out of Taiwan showed that using ECG AI-based alerts to warn physicians about high-risk patients led to an astounding 31% drop in mortality. The study randomized 15,965 hospitalized patients to either receive ECG AI monitoring or not, alerting physicians if the AI found that patients were at high risk of mortality. Patients in the ECG AI group had far lower rates of all-cause mortality (3.6% vs. 4.3%) and cardiac death (0.2% vs. 2.4%) after 90 days.

• Starlight’s CHD Grant: Starlight Cardiovascular scored a $2M grant from the NHLBI and the NIH to support preclinical testing of its Pulmonary Artery Flow Restrictor devices, which are intended to prevent over-circulation to the lungs in babies with Congenital Heart Disease. Starlight’s restrictor devices could serve as a minimally-invasive alternative to open-heart surgery, as they are delivered via a small catheter and can be expanded as the babies grow. The grant comes about 18 months after Starlight completed its $9.5M Series A. 

• TAVR>SAVR? Not So Fast: A review of RCTs comparing SAVR to TAVR highlights the need for multiple statistical methods and larger samples to assess TAVR’s risks and benefits before wider clinical adoption. Utilizing trial sequential analysis (TSA) across 8 RCTs, researchers found no significant reduction in one-year mortality for TAVR in higher-risk trials (RR=0.93). Lower-risk trials showed a mortality difference in conventional meta-analysis (RR=0.67, P=0.031), but TSA indicated potential spurious evidence (P=0.116), stressing the need for more data to confirm TAVR’s benefits.

• CVAUSA Acquires in Phoenix: Cardiovascular Associates of America (CVAUSA) expanded its presence in the Southwest, partnering with Phoenix-area practice Atria Heart (14 physicians, 4 locations). The acquisition gives the private-equity backed platform three practices in Arizona, and CVAUSA called Atria Heart a “huge milestone” for its “Arizona statewide strategy.” Atria is CVAUSA’s third practice acquisition of 2024 (the previous two weren’t announced), after acquiring at least 10 practices in 2023 and assembling 21 groups since 2021.

• LAAE’s Non-AF Benefits: A study of patients without AFib who underwent open cardiac surgery (coronary bypass or valve surgery) found that the patients who also received left atrial appendage exclusion (LAAE) had reduced four-year risks of all-cause mortality (9.9% vs. 13.4%) and ischemic stroke (12.0% vs. 16.0%). While LAAE seems logical for higher-risk non-AF patients due to clot formation concerns, recipients did experience higher rates of postoperative atrial tachyarrhythmia (13.6% vs. 6.8%) and oral anticoagulant usage (24.8% vs. 17.0%).

• Apple’s Trial-Ready AFib App: The Apple Watch’s atrial fibrillation history feature just became the first digital biomarker approved for use in clinical trials, after its addition to the FDA Medical Device Development Tools (MDDT) program. The Apple Watch AFib history feature reports estimated AFib burden, and is now qualified for use as a secondary efficacy endpoint in cardiac ablation clinical trials, while potentially paving the way for the use of other sensor-based digital health technologies in clinical trials.

• SGLT2is Ease Mitral Regurgitation: New data support the use of SGLT2is for easing functional mitral regurgitation in heart failure patients with mild-to-moderately reduced ejection fraction. Among 128 patients in the EFFORT study, the SGLT2i ertugliflozin (Merck/Pfizer’s Steglatro) surpassed placebo in reducing three measures: regurgitant volume (-11mL), left atrial volume index (-6 mL/m²) and LV longitudinal strain (-1.44%). Rates of serious adverse events were also numerically lower for ertugliflozin (one event vs. six for placebo).

• FDA to Oversee Lab Tests: The FDA finalized a new rule to treat laboratory-developed tests as medical devices, putting them in the crosshairs of federal oversight. Lab-developed tests have historically been considered low risk, but the new rules will phase in FDA oversight over the next four-years to ensure their results can be trusted. An estimated 12,000 labs will need to be submitted for evaluation, which has led to pushback from hospitals saying the high costs of regulation will only limit patient access.

• AI Combo Predicts Cardiovascular Mortality: A Nature study showed that Cedars-Sinai’s AI-based CAC scoring solution combined with TotalSegmentator open-source segmentation software was able to predict mortality using chest CT scans. The combo had an AUC of 0.69 for all-cause mortality and 0.77 for cardiovascular mortality using CTs from 24k patients in a lung cancer screening dataset, while achieving similar results with two other patient groups.

Experience the Future of Learning: Medtronic Academy 2.0 is Here!

Unlock your ultimate destination for structural heart medical education with the newly redesigned Medtronic Academy 2.0. Gain access to expert-led courses, webinars, and a wealth of resources to stay ahead in cardiovascular care. Visit now!

AHA Honors Stanford & Bunkerhill’s Incidental CAC Paper

Stanford University’s paper exploring how Bunkerhill’s incidental CAC algorithm increased statin use won the American Heart Association’s 2023 Willerson Award for the best clinical paper published in Circulation this year. See how Bunkerhill’s incidental CAC impacted patient care here.

The Resource Wire

• Revolutionizing Heart Attack Prevention: Cleerly founder and CEO Dr. James Min talks with preventive cardiologist Dr. Erica Jones about Cleerly’s groundbreaking new pathway for heart attack prevention. Watch their conversation to learn about Cleerly’s impact on cardiac health outcomes.

• PIA’s Post-Processing Solution: Advanced cardiac imaging often calls for a time-consuming post-processing step, requiring costly software, hardware, and training. See how PIA provides this post-processing at lower cost, improved consistency, and greater efficiency. 

• Explore the Potential of Cardio AI: Explore the potential of AI-powered cardiology solutions in this on-demand TeraRecon webinar, detailing how its Cardio Suite solutions help expedite disease diagnosis, care coordination, and provide the data to support cardiac treatment decisions.

• Echo AI for HFpEF: New research comparing Us2.ai to gold-standard invasive hemodynamic measurement showed that in patients with HFpEF, echo AI measurements are interchangeable with manual core-lab measures to diagnose increased filling pressures – and could improve HFpEF detection.