In its use artificial intelligence for the early detection of vaginal fibrillation and the prediction of strokes, modern cardiological research is directed.
"Our attention in recent years has focused on how artificial intelligence and wearable devices can help in early detection and prevention, aiming to identify individuals at increased risk," notes Reader of Personalized Cardiovascular Medicine at the Third University Heart Hospital AUTH, at Hippokratia Hospital, Constantine Bakoyannis, speaking at the Athenian-Macedonian News Agency, on the margins of the conference "Happocrates Days of Cardiology 2026".
The conference, held 14 and 14 May in Thessaloniki, was organized by the Second University Cardiological Clinic of ATH in collaboration with the Atherosclerosis Society of Northern Greece.
The silent danger of vaginal fibrillation
As Mr Bakoyannis explains, there are patients who manifest, many times, small or even larger strokes without apparent underlying cause or some cardiological or other condition that justifies it.
"A light stroke can be treated relatively easily. But if it is heavier, it has implications that significantly affect functionality and quality of life. One of the most important "secret" risk factors is vaginal fibrillation, arrhythmia in which the vaginas of the heart enter chaotic function, favoring the formation of clots that can reach the brain and cause ischemic stroke. We know that 25%-30% of strokes are due to "silent" vaginal fibrillation, he notes, stressing the severity of the condition, which often causes no noticeable symptoms.
Particularly difficult to diagnose is paroxysmic vaginal fibrillation – the most common form of arrhythmia – in which episodes appear and retreat automatically within 24-48 hours. An official diagnosis requires a recording of at least 30 seconds of continuous arrhythmia, which is not always possible with conventional methods, as the episode can manifest even once every few months.
Algorithms that "see" what the human eye loses
Aiming at identifying people with increased risk and early prevention, research turns to the use of artificial intelligence.
"The artificial intelligence helps us on many levels. One of the basics is that we can, even from a normal electrocardiogram, identify which individuals are more at risk of developing vaginal fibrillation. This is because, before arrhythmia occurs, there are minor structural changes in the heart, such as enlargement of the sinuses. These changes are so subtle that the human eye hardly detects them," he says.
"With the use of artificial intelligence algorithms, our predictive ability improves and we can identify who has greater risk, even without symptoms," he adds, pointing out that in the same direction the wearable devices are moving as well as modern pressure meters with the ability to detect arrhythmias.
Anticoagulant drugs are the basic "sign" against strokes in patients with vaginal fibrillation, but increase the risk of bleeding. As Mr Bakoyannis notes, balancing benefits and risks for each patient separately is exactly the field where personified medicine, with the help of artificial intelligence, can prove decisive. "The artificial intelligence can also help us identify which patient benefits more from a cure and contribute to its individualization," he adds.
Research at an experimental stage
Bakoyannis points out that it cannot be predicted with certainty when artificial intelligence will join the daily clinical practice, as most algorithms are still at an experimental stage, with studies carried out mainly in small populations. In order to confirm their reliability and safety, extensive research is required in larger and different populations.
"Whether all of this will soon join the daily clinical practice depends on many factors: data availability, regulatory framework and overall organisation of the health system. If such tools are implemented massively, they can also lead to excessive use of health services, with more people worrying and addressing doctors more frequently. All these issues are under constant discussion at European and international level. Therefore I remain restrained as to when we can say that such tools are part of daily clinical practice," he explains.
The Third University Cardiac Clinic, pioneered by Professor Vassilis Vassilikos, has been active in this field for over ten years. Today he is conducting studies in patients with paroxysmic vaginal fibrillation and in patients after catalysis, with a view to predicting recurrence, in collaboration with Harvard and New York universities, to check the effectiveness of algorithms in different populations.