Taking a tour of even some of the historical branches of coronary artery disease (CAD) research, should make any healthcare professional excited in anticipation of the synthesizing effect of emerging technologies, such as AI, in CAD diagnosis, treatment and prevention.

History of Coronary Artery Disease Research and Diagnosis seems to predict a great leap forward

Coronary artery disease (CAD) is a leading cause of morbidity and mortality worldwide. While often thought of as a “modern disease”, Coronary Artery Disease is probably as old as humanity. In evolutionary terms, the complex effects of prolonged lack of oxygen to myocardial tissue - as a result of the reduced blood flow via the arteries that supply the heart itself - are likely as old as the structure and function of any hominid heart.

Long before the terms "atherosclerosis" or "ischemia" were coined, humanity had already begun encountering the symptoms of coronary artery disease. Ancient Egyptian mummies have shown evidence of arterial calcification, suggesting that atherosclerosis – the buildup of atheroma within blood vessels – has existed for millennia. These calcified plaques, visible through modern imaging like CT and MR contrast-enhanced scans, tell a story of a disease that spans human history.

The history of diagnosing and treatment of CAD is a fascinating journey reflecting advancements in medical understanding and technology. As is often the case, the various innovations serving as milestones on this journey say as much about the age in which they were introduced as they do about the disease they are attempting to diagnose or treat.

For most of humanity's evolution and more recent sentient, recorded history, just from symptoms observed, the morbidity related to the functional “design” of our heart’s own plumbing must have been a terrifying (and perceived mostly as mind-bogglingly random) way to go. “If only we could recognize the signs earlier, predict the “attack” from early occurring symptoms” ...maybe then, we could prevent the seemingly random, sudden deaths resulting.

This became the mantra, even the obsession of physicians and researchers focused on solving this puzzle. Starting from very caring, but only anecdotal observation, the inquisitive, but slow pattern-searching followed. Different branches of research began to take shape. Then, as in every other area of medical research, art gave way to science and through experimentation and innovation we started to make progress in researching the heart and the role of the coronary vessels as well. However, while each branch built on the conclusions of previous studies, the different areas of research have remained largely independent in approach and as a result, often stayed isolated in their progress as well.

From Early Clinical Observations to Linking Symptoms to the Heart - The age of enlightenment and the honing of the scientific process:

Historical medical texts, including those from ancient Greek and Chinese medicine, documented symptoms like chest pain, shortness of breath, and fainting – signs now often associated with CAD. While these early observations were often interpreted through mystical or humoral lenses, they laid the foundation for the clinical curiosity that would fuel centuries of cardiovascular research.

Before any specific diagnostic tools existed, CAD was primarily recognized through its symptoms, most notably as chest pain. During the renaissance, anatomical discoveries through autopsies started to uncover the links between the structure of the heart and blood supply but the cause of angina remained too complex for the models of the age.

William Heberden provided a detailed clinical description of angina pectoris in the late 18th century, linking it to exertion and emotional stress – further complicating the observed model. However, the underlying cause of narrowed coronary arteries wasn't fully understood.

The Dawn of Objective Diagnosis (Early 20th Century):

Early 1900s: Electrocardiogram (ECG):The excitement of the age about all things electric, predictably, was applied to cardiac research as well. The early 20th century introduced a transformative leap in cardiac diagnosis with the invention of the electrocardiogram by Willem Einthoven. This groundbreaking device allowed doctors to record the heart’s electrical activity - a revolutionary method that marked the first truly objective diagnostic cardiac exam. ECG allowed physicians to record the electrical activity of the heart and identify patterns indicative of myocardial ischemia and infarction. It also signaled the shift from just understanding to actually trying to do something about cardiac diseases.

Visualizing the Arteries - Catheterization and Coronary Angiography:

• The magic of looking inside the human body without having to take it apart, through Wilhelm Roentgen’s X-rays, was obviously going to be aimed at the heart as well. The moment the technology moved from the hands of WWI military field-medics to routine practice in hospitals, the first cardiac fluoroscopy studies were also performed.
• The bizarre, self-experimentation story of first the cardiac catheterization was practically inevitable and also reflected the weirdness of the age between the world-wars. But, as Forssmann walked the stairs from the OR to the x-ray room a floor below with a catheter he inserted into his own right ventricle via his antecubital vein in 1929, (just wow!!) the practice of cardiology changed forever– the stunt proved that internal cardiac access was possible.
• The development of coronary angiography by Mason Sones by the late 1950s logically followed and, indeed, marked another significant leap in our understanding. The concept of injecting a contrast dye into the coronary arteries and taking X-ray images, allowed direct visualization of arterial blockages in real time and in-situ mapping of coronary arteries and their stenosis. Angiography quickly became the "gold standard" for definitively diagnosing the presence and extent of CAD, further improving our model of not only how a normal human heart tissue is structured and supplied with oxygen, but also showing how narrowing's and blockages in “this plumbing” result in the disease that can lead to potentially fatal clinical consequences. While our functional modeling and understanding of the disease took millennia, it took Dr. Sones less than a decade to move from the first coronary angiogram to the first coronary artery bypass graft (CABG) procedure.

Expanding Diagnostic Capabilities in the late 20th and early 21st Centuries:

• Now that we had a clear model of CAD with a viable “fix”, the search was on to recognize it earlier, with decreasingly invasive procedures. The objective became to make the diagnosis more definitive and more predictive, while also reducing the risk from the diagnostic procedures themselves. At the same time, finding new patterns now visible from the combination of deploying the different diagnostic modalities that became available increased our understanding of the cause and the clinical risk from the extent and location of the stenosis. 
• Stress Testing: Exercise electrocardiography (stress testing) gained prominence as a non-invasive way to provoke symptoms and ECG changes suggestive of CAD during physical exertion, combining clear anatomical understanding of heart function with the patterns of electrical activity.

Prominence of Imaging:

• One unexpectedly positive legacy of the “nuclear age” – the medical use of isotopes for functional imaging was predictably applied to CAD diagnosis as well. Nuclear Cardiology techniques like myocardial perfusion imaging using radioactive tracers emerged to assess blood flow to the heart muscle under rest and stress conditions. 

The recognition of the negative effects of radiation necessitated the innovation of imaging while minimizing ionizing radiation. Ultrasound imaging of the heart became a valuable tool to assess heart function and sometimes visualize signs of ischemia. Combining ECG with ultrasound, stress echocardiography further enhanced CAD diagnostic capability. 

• Computed Tomography Angiography (CTA): As the age of analog instrumentation gave way to the age of digital signal processing and computers, CT imaging was born. In Cardiology, non-invasive, Computed Tomography Angiography(CTA) logically followed. In more recent decades, even though it is using x-rays, non-invasive CT angiography has become increasingly sophisticated, providing detailed 3D images of the coronary arteries without the need for catheterization in many cases. The near-ubiquitous access to CT scanners made cardiac CT (CCT) studies very common. The acquisition protocols have also evolved to better visualize lipid and calcium deposits, while quantitative analysis and standardized reporting of CAD continues to hold further great potential.
• Cardiac Magnetic Resonance Imaging (CMR): Along with CCT, Cardiac MRI offers detailed information about heart structure and function and can be used to detect myocardial scar tissue and assess blood flow. Being a non-invasive, radiation-free imaging modality, CMR imaging is useful for the management of patients with CAD. Various MR acquisition techniques and protocols have been developed over the last three decades to evaluate cardiac function and detect defects in myocardial perfusion. Specifically, late gadolinium enhancement (LGE) imaging, a well-established technique that uses contrast to highlight scar tissue, can identify the presence and extent of scar tissue in the heart, which is a common finding in CAD. While CMR has a high degree of accuracy and reliability in detecting and characterizing CAD, the modality also allows accurate risk stratification of patients with established CAD.
• Post-processing and Reporting: While the amount of imaging studies and the corresponding imaging data has increased exponentially, the need for post-processing has also sky-rocketed, making automation necessary. Machine-learning techniques make it increasingly possible to minimize the manual effort required to not only accurately map the cardiac structure, function and blood-flow but to also to quantitatively analyze cardiovascular diseases, including CAD. Structured reporting’s adoption in cardiology is significantly ahead of other disciplines. Yet, only in the last decade have we started routinely categorizing and stratifying CAD, based on the extent of stenosis and overall plaque burden. Needless to say, quantified analysis and standardized reporting represent both a great downstream clinical value and a gold-mine of data for population-level analysis.
• Biomarkers for CAD: Another great branch of CAD research has been producing deeper understanding by identifying and utilizing various proteins, peptides and enzymes that are involved in CAD. Troponins, BNP, CRP, MPO, Lp(a) and IL-6– to name a few - each play a unique role in the diagnosis, risk assessment, and management of CAD. Troponins are considered the gold standard for diagnosing myocardial infarction (MI) and are highly sensitive and specific for cardiac injury. Elevated troponin levels can indicate ongoing ischemia and are used to assess the severity of CAD, often guiding decisions in emergency settings, triaging patients for further imaging or intervention. As research continues to evolve, these and other biomarkers may offer new insights into the pathophysiology of CAD. Labs now use highly sensitive assays to detect even minute changes, and AI-powered platforms hold the potential to synthesize this multifactor, often unstructured data into actionable clinical insights.
  
• Genetics and epigenetics - Unlocking Hereditary Risk: Recent studies have identified numerous genetic variants associated with CAD. Genome-wide association studies (GWAS) have pinpointed specific genes linked to lipid metabolism, inflammation, and vascular function. Some of these notable genetic markers associated with CAD have been studied extensively, including:
• PCSK9: Variants in this gene can lead to elevated cholesterol levels, increasing CAD risk.
• LDLR: Mutations in the LDL receptor gene are linked to familial hypercholesterolemia, a condition that significantly raises the risk of CAD.
• APOE: The APOE gene is involved in lipid metabolism, and certain alleles are associated with increased CAD risk.

Ongoing studies aim to discover more genetic variants and their functional implications. Additionally, advancements in gene editing technologies, such as CRISPR, hold promise for developing novel therapies that target the genetic basis of CAD. By leveraging genetic insights, healthcare providers will be empowered to routinely offer personalized care that addresses the unique risk factors of each patient. 

Present-Day Landscape: Merging Traditional and Tech-Driven Tools

By now, in the early 21st century, we understand that Coronary Artery Disease occurs when the coronary arteries become narrowed or blocked, often due to atherosclerosis. Statistically, we have identified known risk factors, that include high cholesterol, hypertension, smoking, and diabetes. Nevertheless, CAD is a complex disease that varies in presentation and its exact progression in an individual remains difficult to predict. The biological mechanisms that lead to lipid deposits and calcification leading to the progression of the disease are relatively well understood. The less deterministic impact of family history, genetics and epigenetics is much less understood.

Today, the diagnosis of CAD often involves a combination of the previously mentioned methods, tailored to the individual patient's symptoms, risk factors, and clinical presentation. Simple, diagnostic stress-ECG tests are most often used for initial evaluation followed often by modern, non-invasive imaging studies, such as an echo and cross-sectional imaging studies (CCTA or CMR), with coronary angiography reserved for cases requiring more definitive diagnosis or intervention. The field continues to evolve with ongoing research into new imaging and other diagnostic techniques, with each test carrying different advantages and disadvantages. 

Progress needs AI: The Case for Integration

Overcoming Fragmentation in Research and Practice

Progress is never a linear progression. Just like in other areas of cardiac research, and indeed all of medicine, of course, promising work is being done continuously in independent, but often diverging, sometimes even isolated areas of research. The practice of medicine organized into compartmentalized healthcare departments brings focus to sub-specialties, but, combined with the funding structure through grants, the deep and narrow areas of research often continue in silos without the benefit of broader perspectives. This fragmentation not only creates frustrating confusion for patients, it also limits healthcare’s ability to generate holistic, personalized care plans.

It is high time to bring our historical learnings together – converging them, breaching conventional boundaries to improve cardiac research and care. Just like in other areas of life, we need to work to figure out how to bring AI to play a key role in integrating huge volumes of data and synthesize the information within various adjacent but seemingly independent scientific domains.AI doesn’t just process data- by analyzing diverse and large datasets, AI could potentially help predict coronary artery disease risk, develop personalized interventions for individual patients, effectively synthesizing knowledge from various biological, diagnostic and medical disciplines.

AI also holds the potential to assist researchers by generating novel research hypotheses, detailed overviews, and experimental protocols based on a specified research goal for CAD research. AI could and should be harnessed to improve clinical practice diagnosis and subsequent management. 

Conclusion

From the earliest symptom observations to today’s cutting-edge diagnostics, each era has added layers to our understanding of CAD. Can't wait to see what yet unpredictable benefits the age of AI will bring to the deepening of our understanding, assessment ultimately to the treatment of this pervasive condition.

It is exciting to imagine what the ever-expanding pattern-search, pattern recognition, and modeling capabilities of even narrow-AI algorithms will bring to the table in terms of personalized risk assessment from multifactor quantified analysis, early screening, based on genetic and epigenetic risk factors, incidental findings from non-cardiac intended chest studies, and more.

Maybe, just maybe,…we are on the cusp of bringing many possible angles of research, population health and diagnostic techniques together to make them quantified enough to bring these new, learning, modeling, intelligent tools to bear early enough for tackling CAD prevention- not just for an individual patient but for an at-risk population segment.

To learn more about how AI-based CCTA post-processing can be part of your practice, visit www.circlecvi.com/cardiac-ct.

As research continues to evolve, the integration of the many different “branches of research” into clinical practice supercharged by AI, will undoubtedly enhance our ability to prevent, diagnose and treat CAD, ultimately improving patient outcomes and reducing the burden of this prevalent disease.

Frequently Asked Questions (FAQs)

Q1. What are the most frequently used tests for diagnosing coronary artery disease?

A combination of stress testing, cardiac CT angiography (CCTA), cardiac MRI, and biomarkers like troponin offers the highest diagnostic accuracy. Each test has unique advantages depending on patient risk profile.

Q2. How can artificial intelligence improve CAD diagnosis?

AI has the potential to improve CAD diagnosis by analyzing imaging, ECG, and lab datasets faster and with higher accuracy and consistency. It can help detect patterns in multi-modality datasets in the longitudinal health record of an individual patient, or can be an invaluable tool for analyzing large, multi-factor datasets across many patient groups. 

Q3. What role do genetics play in CAD?

Genetics significantly affect CAD risk. Variants in genes like PCSK9, LDLR, and APOE can influence cholesterol metabolism, inflammation, and vascular function. Epigenetics further modify gene expression in the risk of CAD for a given patient.

Q4. Can CAD be detected early without invasive procedures?

Yes. Non-invasive imaging techniques like CT angiography, cardiac MRI, and stress testing, combined with biomarker analysis, can detect early signs of CAD without invasive catheterization. Given the general pervasiveness of CT for many different type of trauma or other clinical reasons for abdominal/chest CT acute studies, including additional quantitative analysis of CAD risk might carry significant value in the long run.

Q5. What’s next for CAD diagnostics?

Future diagnostics will rely heavily on AI, predictive modeling, and integration of multi-modal data including genomics, imaging, and lifestyle tracking. These advancements aim to enable early detection and prevention.