Sudden cardiac death is one of the leading causes of death in Western countries. An estimate of the annual global incidence of sudden cardiac death is around 4 to 5 million cases per year. In the particular case of sudden death in sport, it is estimated that in Spain it occurs in 2 out of every 100,000 people under 35 years of age (some 350 deaths annually among young people) and 1 out of every 18,000 in the 25 to 75 year age range.
Recognising the possible causes of sudden death is a major challenge today. The first step in trying to reduce its incidence involves a medical examination prior to participation in sport. With this initiative it is possible to detect some of the causes that may lead to these fatal events. However, the existence of a considerable number of patients who suffer sudden death as their first symptom represents a major constraint in solving this problem.
Paradoxically, there are more and more sports fans and more sports events without an increase in the number of cardiological examinations. In this scenario it is essential to develop advanced tools that incorporate demographic data, medical data and results of sensitive cardiological examinations that facilitate the work of the cardiologist, leading to the stratification of sportsmen and women according to their cardiovascular risk.
Artificial intelligence, particularly the use of advanced algorithms based on machine learning, represents an opportunity in this field. The analysis of complex statistical patterns combined with BigData techniques will allow the development of new predictive models of value for professionals in the field of cardiology.
In this sense, and as a result of our continuous activity in open innovation, in 2020 we will start a collaboration with LetsCardio, a digital platform that groups together sports cardiology centres from all over Spain, and Innatial Developers, an expert software consultant.
With this collaboration we intend to combine knowledge in the field of sports cardiology and the application of machine learning techniques, trying to identify and discover hidden patterns in the data to improve diagnoses and prevent fatal events.