SUDDEN cardiac death remains the leading medical cause of death among young athletes, and new research suggests that explainable artificial intelligence (AI) combined with advanced digital heart models could help identify those at greatest risk before a life-threatening event occurs.
Athlete Cardiac Safety Back in the Spotlight
Public attention on athlete cardiac safety was renewed this week after Denmark footballer Christian Eriksen collapsed during a friendly match against Ukraine when his implantable cardioverter defibrillator delivered a corrective shock, highlighting the ongoing challenge of managing potentially dangerous heart rhythm abnormalities in sport.
The incident underscored a broader question facing sports medicine: how can clinicians better identify athletes at risk of serious cardiac events before they occur?
A new systematic review sought to address this challenge by examining the potential role of explainable artificial intelligence (AI) and advanced cardiac modelling in athlete screening.
Sports-Related Sudden Cardiac Death Remains a Challenge
Researchers conducted a systematic review of studies published between 2013–2025 examining sports-related sudden cardiac death, explainable AI approaches for detecting life-threatening arrhythmias, and cardiac electrophysiological models.
Of 9,574 studies identified, 84 met the inclusion criteria, including 16 epidemiological studies, 30 focused on explainable AI, and 38 investigating cardiac modelling. Across the epidemiological studies, the incidence of sports-related sudden cardiac death ranged from 0.1–0.6 cases per 100,000 participants annually.
Although rare, these events remain a significant concern because they can occur without warning and have devastating consequences for athletes, families, and sporting communities.
Explainable AI and Digital Hearts
The review found that Gradient-weighted Class Activation Mapping was the most commonly used explainable AI technique. Unlike conventional AI systems that can be difficult to interpret, these approaches help clinicians understand which features influence a prediction, potentially improving confidence in AI-assisted decision-making.
Researchers also examined cardiac electrophysiological models that simulate the heart’s electrical activity. Most focused on cellular and tissue-level processes involved in arrhythmia development.
The authors suggest that combining explainable AI with these digital heart models could support more personalised risk assessments and improve the identification of athletes who may be vulnerable to life-threatening cardiac events.
Towards Better Athlete Screening
The review highlighted substantial variation in how sports-related sudden cardiac arrest and sudden cardiac death are defined and reported across studies.
The authors argue that standardised definitions and better integration of epidemiological risk factors, AI tools, and cardiac modelling frameworks will be essential to advance athlete-specific risk stratification.
Such advances could ultimately help clinicians identify high-risk athletes earlier and strengthen efforts to prevent catastrophic cardiac events before they occur.
Reference
Vanegas Müller E et al. A systematic review of explainable artificial intelligence and cardiac electrophysiological models addressing sports-related sudden cardiac death and arrest in adolescents and young adults. npj Digit. Med. 2026;DOI:10.1038/s41746-026-02878-x.
Featured image: Hamza on Adobe Stock
- Author:
