The National Heart, Lung, and Blood Institute (NHLBI) and the National Institute of Biomedical Imaging and Bioengineering (NIBIB), of the National Institutes of Health (NIH), convened an in-person workshop in Bethesda, Maryland, on September 17-18, 2025, titled "Emerging Technologies for Coronary Artery Disease Imaging: Vision 2040."  The goal of the workshop was to identify research opportunities that will shape the future of Coronary Artery Disease (CAD) screening and prevention through advances in imaging.

Background:

Cardiovascular disease remains the leading cause of death in the United States, with coronary artery disease (CAD) (e.g., ischemic heart disease) accounting for over half of deaths, underscoring the critical need for tools that can detect CAD before symptoms appear. Recent data suggest that 40-45 year-old individuals have measurable plaque. Furthermore, significant plaque burden is associated with significant risk of major adverse cardiovascular events (MACE). In contrast to mammography for breast cancer, CAD is often only detected after clinical events occur in the most at-risk individuals.

Although existing imaging modalities such as Computed Tomography (CT) coronary artery calcium scoring, CT angiography, MRI, Positron Emission Tomography (PET) and Single Photon Emission Computed Tomography (SPECT), echocardiography have improved risk stratification, they remain limited in accessibility, cost, radiation exposure, and validation for population-level use. In practice, the wide variety of diagnosis modalities and lack of standardization also make it difficult to effectively utilize existing data acquisition for early diagnosis and prevention. Advancing CAD detection and prevention aligns with the broader public health imperative of the NHLBI strategic vision to translate discovery science into clinical applications that improve outcomes across populations.

Summary of Discussion:

The workshop brought together leaders and experts in cardiovascular imaging, engineering, clinical research, and artificial intelligence (AI) to identify research opportunities for developing accessible, efficient screening tools to detect subclinical Coronary Artery Disease (CAD) and reduce CAD-related morbidity and mortality by 2040. Participants examined emerging technologies, including advances in CT, Magnetic Resonance Imaging (MRI), ultrasound imaging, molecular imaging, and AI image processing, to assess their potential to enable a safe, widely-accessible screening tool for subclinical CAD. Discussions focused on scientific and engineering innovation, validation frameworks, clinical deployment pathways, and opportunities to shift cardiovascular care toward earlier detection and prevention at the population level, while also focusing on personalized medicine, since population level statistics often cannot effectively inform individual patient treatments, outcomes, or clinical recommendations.

Knowledge Gaps and Opportunities Identified from Discussions:

Figure 1: Timeline of predicted Coronary Artery Disease Imaging timeline with additional insights/events added during the proceedings of the workshop.

1. The Role of Imaging in Screening and Prevention

Gaps:

• Imaging for CAD mainly targets severe stenosis, leading to late-stage, intensive treatment, while early atherosclerosis is not routinely screened.
• Many CAD screening technologies have limited access and adoption due to cost, radiation, and workflow barriers.

Research Opportunities:

• Build a mammography/colonoscopy-style CAD screening program that is safe, affordable, workflow-supported, and effective across diverse populations and settings, including global and community use.
• Use scalable, low-cost imaging to detect subclinical disease early and enable longitudinal monitoring, even when coronary calcium is zero.
• Expand beyond anatomy to biology-informed imaging (inflammation/cellular activity) using tools like advanced ultrasound and compact PET to support personalized prevention and early-life adherence/education.

2. Characterizing CAD Pathophysiology

Gaps:

• Sex differences are underrecognized: many women have angina and adverse cardiovascular events despite minimal anatomic disease and no Standard Modifiable Risk Factors (SMuRFs).
• CAD is still focused mainly on flow-limiting plaque, rather than taking into account the biological/inflammatory drivers of the disease
• Key plaque injury types (rupture or erosion) are not consistently distinguished, limiting understanding of vulnerability.

Research Opportunities:

• Emphasize early-life metabolic risk (LDL, blood pressure, triglycerides) as a major driver of later CAD and a rationale for prevention.
• Expand the model to include inflammation and systemic pathways (immune activity, bone marrow activation, microvascular dysfunction, brain perfusion changes) that drive vascular aging and malignant atherosclerotic progression.
• Predict events using vulnerability biology and plaque features such as plaque burden, lipid-rich plaques, remodeling/necrotic core, and lumen size to treat both microvascular and obstructive disease as a continuum, recognizing myocardial infarction (MI) can occur via multiple mechanisms beyond stenosis.

3. CT Technologies for CAD Risk Identification

Gaps:

• Calcium scoring and ischemia-only pathways are insufficient, missing important drivers of MI risk.
• Quantitative Coronary CT Angiography (CCTA)lacks standardization across scanners/protocols, limiting reproducibility.
• High-risk plaque is often detected too late (after MI), increasing downstream CAD complications.

Research Opportunities:

• Use CCTA (plus complementary imaging when needed) to detect obstructive and nonobstructive plaque, identify high-risk features early, and potentially screen select asymptomatic individuals to improve risk stratification beyond calcium alone.
• Add biology and physiology to CT risk models. Some options discussed include Fat Attenuation Index (FAI) for vessel-specific inflammation, CT perfusion for physiologic assessment, and CT-derived Fractional Flow Reserve (FFR) to link anatomy/plaque with functional risk.
• Improve and standardize CT for consistent quantitative results. Discussions mentioned photon-counting CT, better detectors/motion correction, and AI analytics, while pushing cross-scanner/protocol standardization.

4. Evaluation and Integration of Emerging Technologies

Gaps:

• CAD screening/diagnostic tests are fragmented and not integrated, leading to inconsistent interpretation, unnecessary interventions, and missed high-risk plaque features.
• Wearables and mobile sensors are underused, and current imaging is treated as single snapshots rather than tracking disease over time.
• There’s limited data infrastructure and usability: few high-quality linked imaging databases currently exist, and equipment is complex and requires highly trained operators.

Research Opportunities:

• Create unified, streamlined imaging pathways that combine fewer modalities and integrate physiology and plaque characterization (e.g., perfusion/flow metrics; intravascular tools where appropriate and better noninvasive alternatives).
• Integrate advanced imaging, sensors/wearables, smartphones, and AI to expand screening access, improve patient engagement, and enable longitudinal monitoring.
• Build high-quality, multimodal imaging data platforms (linked to clinical records) to support AI development, bias assessment, regulatory-grade datasets, and develop simpler, guided devices usable by non-specialists in clinics and communities.

5. Implementation Strategies and Clinical Trials

Gaps:

• CAD screening deployment lacks a clear plan: needed tool features, validation, access models, pilots, and partnerships aren’t well defined.
• Existing trial data and ongoing studies that could support CAD screening and imaging validation are not being fully leveraged.

Research Opportunities:

• Create a CAD Imaging 2040 roadmap that aligns imaging science, engineering/computation, trial design, regulation, and real-world implementation.
• Use existing large trial infrastructure and nested imaging sub-studies to validate imaging-driven screening/monitoring, define surrogate endpoints, and run pragmatic prevention trials with standardized imaging and clinical outcomes.
• Build imaging-informed care pathways and biomarkers (e.g., inflammation measures like FAI) to guide prevention, personalize therapy, and evaluate how treatments (e.g., statins, GLP-1 agents) change plaque and progression.

Conclusion:

The CAD Imaging 2040 workshop highlighted persistent gaps in CAD screening and prevention that limit early detection of subclinical disease and the ability to tailor prevention strategies. These gaps define a set of research opportunities focused on advancing imaging-enabled, biology-informed, and population-scalable approaches to CAD risk identification.

Key research opportunities for the field/investigators:

• Define the essential technical, biological, and clinical requirements for a next generation, widely-accessible CAD screening tool capable of detecting early, nonobstructive, and biologically active disease.
• Explore integration of longitudinal multimodal data and frameworks to determine risk stratification and monitoring for early non-symptomatic CAD populations.
• Outline a validation strategy to measure and capture CAD biology (e.g., plaque burden, phenotype, inflammation, and physiologic dysfunction) and their relationship to clinically meaningful risk.
• Explore methods to develop and evaluate standardization of protocol, expansion of screening accessibility (e.g., lower cost, integration into workflow, feasibility), and harmonization of existing multi-modal data frameworks to benefit ongoing research.

Disclaimer

This summary reflects the individual opinions, ideas, themes, perspectives, and opportunities shared by workshop participants. It does not represent consensus recommendations or formal guidance from the NHLBI, NIBIB, or the NIH.

List of Attendees:

Organizers:
Songtao Liu, M.D., Co-Chair, NHLBI
John Haller, Ph.D., Co-Chair, NHLBI
Martin Tornai, Ph.D., NIBIB
Li-Yueh Hsu, D.Sc., NHLBI
Ahmed Hasan, M.D., NHLBI
Youjoung Kim, Ph.D., NHLBI
Guoying Liu, Ph.D., NIBIB
David Schopfer, M.D., NHLBI
Sujata Shanbhag, M.D., NHLBI
Fernando Bruno, M.D., M.P.H., NHLBI

Co-chairs:
Koen Nieman, M.D., Ph.D., Stanford University
Brittany Weber, M.D., Ph.D., UT Southwestern

Speakers:
Keith Channon, FRCP, FMedSci, University of Oxford, UK
Valentin Fuster, M.D., Ph.D., Mt. Sinai Hospital
Bruce Tromberg, Ph.D., NIBIB
Gina Wei, M.D., M.P.H., NHLBI
Kathleen Fenton, M.D., NHLBI
W. Patricia Bandettini, M.D., NHLBI
Kelley Branch, M.D., M.Sc., University of Washington Heart Institute
Marcus Chen, M.D., NHLBI
Bruno De Man, Ph.D., GE HealthCare Technology and Innovation Center 
Zahi Fayad, Ph.D., Icahn School of Medicine at Mount Sinai
Hector Garcia-Garcia, M.D.,  Ph.D., MedStar CV Research Network
Udo Hoffmann, M.D., M.P.H., University of Colorado
Ella Kazerooni, M.D., M.S., University of Michigan Medical School
David Maron, M.D., Stanford University School of Medicine
Cynthia McCollough, Ph.D., Mayo Clinic
Michael McConnell, M.D., Stanford University
Paul Pinksy, Ph.D., National Cancer Institute
Etta Pisano, M.D., American College of Radiology
Harmony Reynolds, M.D., NYU Grossman School of Medicine
Leslee Shaw, Ph.D., Icahn School of Medicine at Mount Sinai
Charles Taylor, Ph.D., University of Texas at Austin
Todd Villines, M.D.,  University of Virginia
Renu Virmani, M.D.,  CVPath Institute
Kim Williams, M.D., FESC, University of Louisville
David Wilson, Ph.D., Case Western Reserve University
Pamela Woodard, M.D., Washington University  in St. Louis
Robert Barry, Ph.D., NIBIB
Andrew D. Choi, M.D., The George Washington University
Joao Lima, M.D., Johns Hopkins University 
Jagat Narula, M.D., Ph.D., The University of Texas Health Science Center at Houston
Leandro Slipczuk M.D., Ph.D., Albert Einstein College of Medicine
Daniel Sodickson, M.D., Ph.D., New York University
Ken Taguchi, Ph.D., Johns Hopkins University
Richard Thompson, Ph.D., Canon Healthcare USA