Skip to main content

AI applications in aortic pathologies

The aorta and the main pathologies

The aorta is the largest artery in the human body. It starts at the heart, in the left ventricle, and travels upwards (ascending aorta), makes a cross and finally descends towards the abdomen (descending aorta) and then divides into two branches, which are called the iliac arteries. The aorta gives rise to numerous arteries and branches that provide oxygenated blood to the whole body, including mainly – the carotid arteries, which supply blood to the brain, – the subclavian arteries, which supply blood to the upper limbs, – the visceral arteries, which supply blood to the various organs in the abdominal cavity, – the iliac arteries, which supply blood to the lower limbs.

The aorta is thus a large artery and diseases occuring in this vessel can have consequences for the whole body. Various lesions, of varying nature and origin, can occur on the aorta, such as:

  • A dissection: This is a tear in the arterial wall into which blood will rush and detach the leaflets of the arterial wall, thus creating a “false channel”.
  • Aneurysm: This is a loss of parallelism in the aortic wall, resulting in a focus dilation of the aorta. In a very schematic way, the wall of the aorta will be deformed and increased in volume, instead of being a regular cylindrical tube.
  • Stenosis and occlusion: Plaques, most often related to atherosclerosis, develop in the aortic wall and can lead to narrowing or obstruction of the vessel.

Aortic aneurysm in clinical practice

How common is this?

Affecting approximately 4.3% of men and 1% of women over the age of 50, aneurysms of the aorta occur when the arterial wall loses its parallelism, resulting in a focus dilatation. Although aneurysms can develop along the entire length of the aorta, 80% of aneurysms are located in the abdominal aorta (corresponding to the portion below the diaphragm). The dilatation most often takes the form of a spindle (fusiform aneurysm in about 80% of cases), sometimes the form of a sac (saccular aneurysm). Aortic aneurysm is also often defined as a diameter of the abdominal aorta larger than 30 mm or one and a half times its normal diameter.

How serious is this?

Abdominal aortic aneurysm (AAA) is most often asymptomatic and discovered during an imaging test performed for another reason. Although the disease is often asymptomatic, it is not benign as it can develop into complications that can be life threatening. The main risk is that the aneurysm increases in size over time and develop into an aortic rupture. As the aorta is the largest artery in the body, a rupture of this vessel is a life-threatening emergency that must be treated immediately.

What is the cause of AAA?

AAA is most often associated with atherosclerosis. Atherosclerosis is a pathological change resulting in the deposition of plaques in large and medium-sized arteries. Atherosclerosis is responsible of many cardiovascular diseases including coronary disease, carotid stenosis or peripheral artery disease. AAA is therefore often associated with other cardiovascular risk factors such as age, male gender, smoking, dyslipidemia (disturbances in cholesterol and/or triglyceride levels) or high blood pressure. Family history is also a risk factor for AAA. This genetic predisposition is complex, multifactorial and not systematically found. More rarely, some aneurysms are linked to identified genetic diseases, such as Marfan’s disease. These are known as “syndromic aneurysms”.

How is the diagnosis made?

The diagnosis of AAA is made through an imaging examination which allows to visualize the aorta, to evaluate the size and the location of the aneurysm. Various imaging techniques are available and used in practice, including abdominal ultrasound (US), computed tomography angiography (CTA) or magnetic resonance imaging (MRI). Ultrasound is most often used for screening as it is easily performed and non-invasive. CTA is the most commonly used technique to diagnose the AAA, perform pre-surgical planning and follow patients.

How is it treated?

There is currently no drug that has been shown to be effective in treating AAA, and the only curative treatment relies on surgery. AAA repair can be performed using open (conventional surgery) or endovascular procedure. In that case, the surgeon makes percutaneous approaches to the femoral arteries to introduce and mount the stent graft in the aorta. The indications for AAA repair will depend on the assessment of the benefits and risks of the surgery. In practice, the indications for AAA repair mainly rely on the evaluation of the size of the aneurysm and its growth rate, with large and/or rapidly growing aneurysms considered to be at high risk of rupture.

Clinical guidelines for patients with AAA are built by international groups of experts and are based on critical analysis of literature and state of the art in order to ensure the best possible management with evidence-based medicine constantly updated with the latest knowledge.

In the field of vascular diseases, clinical practice guidelines are mainly established by two networks of experts:

In France, the French Society of Vascular and Endovascular Surgery (SCVE) also provides practical information to better understand and appreciate vascular diseases and their treatments.

Our research

Our research team focuses on aortic aneurysms through translational research with the aim to – better understand the mechanisms underlying the development of aneurysms, – better diagnose, treat and monitor the disease. We combine efforts in clinical research, basic sciences and innovative technologies to develop a transdisciplinary approach required to improve patients care (renvoyer à la ref à propos et nos publications). At the Institute 3IA Côte d’Azur, we are particularly interested in the applications of Artificial Intelligence (AI) to improve the management of vascular diseases, including aortic aneurysms.

The contribution of Artificial Intelligence (AI) in aortic aneurysm

Despite major advances in the treatment of cardiovascular disease over the last decades, various challenges remain to further improve the management of patients with AAA. AI is a very broad field that offers the opportunity of developing many applications in medicine including to improve cares for patients with aortic aneurysm (see dedicated chapter). For detailed information on the state of the art in this field, you can consult the following article we have published: https://pubmed.ncbi.nlm.nih.gov/32093909/.

AI to improve medical imaging

There is a growing interest in using AI to improve and facilitate the analysis of medical imaging. Indeed, although they are very useful, currently available commercialized software for imaging are semi-automatic, meaning they require visualization and intervention of trained medical doctors to analyze it. These software programs allow to visualize the vascular system of patients and they are used to measure the size of the vessels in order to order stents adapted to the patient’s anatomy. However, performing these measurements is a tedious and time-consuming task, which require trained specialists, with a risk of variation from one operator to another. One of the challenges of AI is to provide clinicians with new and reliable tools to optimize these tasks to improve accuracy, reproducibility and also to save time that can be fully devoted to patients care. The development of fully automatic software for the segmentation and analysis of medical imaging could thus revolutionize the discipline and provide useful complementary tools for clinicians. In the field of vascular imaging, our team was among pioneers to develop a fully automatic software for CT analysis of patients with aortic aneurysms

For his work on AI applications in aortic aneurysm, Fabien Lareyre, a member of our team, has been awarded a prize by the Académie de Médecine in 2021. This kind of application has an interest in clinical practice, by facilitating imaging analysis and help for pre-surgical planning; but also in research, since it allows to extract detailed and quantitative anatomical data which can then be studied to better understand the disease. Various companies and start-ups working in this field have recently been created in France, but also in Europe, offering great perspectives for using these advanced technologies to improve patients care. Here are a few and non-exhaustive links :

AI to develop aid-decision support

Machine Learning algorithms are designed to learn from data and are able to improve their performance as they are fed with new data.  They can explore very complex associations between data without any a priori assumptions and can enable to identify patterns and develop predictive models.

When a patient is diagnosed with an aortic aneurysm, the crucial question is whether the aneurysm should be operated or whether it is better to refrain from surgery and monitor the patient regularly. In the case of a symptomatic aortic aneurysm, there is an indication for emergency surgical treatment because of the risk of aortic rupture. In the case of an asymptomatic aneurysm, current guidelines recommend surgery based on an assessment of the balance between the risk of aneurysm growth and rupture versus the risks associated with surgery. However, unfortunately, doctors do not have a crystal ball to predict the future and there are very few objective tools available to assess the risk of aneurysm expansion. By bringing in new techniques such as Machine Learning, AI could allow the development of new predictive models to better assess the risk of aneurysm progression based on clinical, biological and imaging data of patients. Several pilot studies have been published showing encouraging results of the use of Machine Learning to assess the prognosis of the disease, by predicting the risk of AAA growth and rupture as well as assessing the risk of post-operative complications.

For detailed information on the state of knowledge in this field, you can consult the following article we have published: pubmed.ncbi.nlm.nih.gov/32093909.

AI thus offers new tools to to support decision-making and propose personalized therapeutic approach adapted to each patient (precision medicine).

Remaining challenges

Although AI offers very promising perspectives for improving care, including the management of aortic aneurysms, research is still in its infancy and essential steps remain before implementing these applications in daily clinical practice. First, the AI-applications need to be evaluated and validated by healthcare professionals to assess their performance, efficiency, and safety of use. AI-derived applications are developed from health data and need to be trained on large volumes of data (big data) in order to be efficient and applicable to all patients. From one center to another and from one country to another, medical data are very heterogeneous as they can be generated by different providers and stored in different formats in several registries and repositories. There is thus a real need of standardization and a need to federate actors of research to build international multidisciplinary collaborations in which AI experts and health professionals work together to develop fair, ethic, responsible and trustworthy AI. Ethics remain at the heart of the development of AI-applications which are subject to very strict regulations, in order to guarantee data protection, respect of privacy and patients’ wishes.

Finally, innovation and development of AI-derived applications require adapted infrastructures as well as competent experts. It can represent a significant economic investment, with potential long-term benefits for health at individual and collective levels.