Fondazione Silvers
Fondazione Silvers
The projects
Neural Networks for Arrhythmic Risk Stratification in Brugada Syndrome
In collaboration with the Pavia University and ICS Salvatore Maugeri, Pavia
Project Description: Background Brugada Syndrome (BrS) is an inherited cardiac channelopathy associated with an increased risk of malignant ventricular tachyarrhythmias and sudden cardiac death in individuals with structurally normal hearts¹. The estimated prevalence in the general population ranges from 3 to 20 cases per 10,000 individuals. The disease is caused by abnormalities of cardiac ion channels, most commonly associated with pathogenic variants in the SCN5A gene, which encodes the cardiac sodium channel Nav1.5. The clinical expression of BrS is highly heterogeneous and ranges from complete absence of symptoms to syncope, polymorphic ventricular tachycardia, ventricular fibrillation (VF), and sudden cardiac death (SCD), which in some cases may represent the initial manifestation of the disease. Although pharmacological therapies and epicardial ablation procedures can effectively control symptoms and reduce arrhythmic burden, the implantable cardioverter-defibrillator (ICD) remains the only proven strategy for the prevention of sudden cardiac death in patients considered to be at high risk¹⁻³. However, ICD implantation is associated with significant complications, particularly in younger patients, including inappropriate shocks, device-related infections, lead malfunction, and the need for multiple device replacements throughout life. Therefore, accurate risk stratification is essential to identify patients who are most likely to benefit from device implantation. Over the last decade, several clinical and electrocardiographic markers associated with arrhythmic risk in BrS have been proposed¹˒⁴˒⁵. Nevertheless, currently available risk stratification models exhibit limited predictive performance and relatively poor positive predictive value. Consequently, some patients classified as low risk may still experience ventricular fibrillation or sudden cardiac death, whereas others may undergo unnecessary invasive procedures. - 2 - This limited prognostic accuracy remains one of the major unresolved challenges in the clinical management of BrS and highlights the need for novel markers capable of improving arrhythmic risk stratification. Enhanced risk prediction would enable more appropriate selection of candidates for invasive interventions while reducing unnecessary procedures in genuinely low-risk individuals. Scientific Coordinators Prof. Carlo Napolitano Dr. Luca Grisorio Objectives The primary objective of this project is to identify electrocardiographic features associated with an increased risk of malignant ventricular arrhythmias and sudden cardiac death in patients affected by Brugada Syndrome. To achieve this goal, machine learning models⁶˒⁷ will be applied to the advanced analysis of 12-lead Holter ECG recordings, with the aim of improving the identification of patients truly susceptible to major arrhythmic events. Expected Outcomes, Scientific Impact, and Translational Potential The project aims to develop and validate a machine learning-based prognostic model for risk stratification in patients with Brugada Syndrome, with the potential to generate significant clinical, scientific, and methodological impact. Specifically, the project is expected to: Improve arrhythmic risk stratification Enhance the accuracy of arrhythmic risk prediction compared with currently available tools, enabling more precise identification of patients at risk of developing malignant ventricular arrhythmias and sudden cardiac death. This may support more personalized therapeutic decision-making, reducing both undertreatment of high-risk patients and unnecessary invasive procedures in low-risk individuals. Establish a comprehensive BrS database Develop a structured and extensively annotated database containing clinical data and 12-lead Holter ECG recordings from one of the largest internationally available cohorts of patients with Brugada Syndrome. This resource will provide substantial value for future studies aimed at improving our understanding of disease natural history, determinants of arrhythmic risk, and factors associated with long-term prognosis. Develop an AI-driven ECG analysis workflow - 3 - Design and validate a complete workflow for the automated analysis of Holter ECG recordings using Artificial Intelligence techniques. This approach may serve as a transferable methodological framework for other inherited and acquired arrhythmic disorders, fostering the development of innovative diagnostic and prognostic tools within the context of precision medicine. Project Duration: Two years.
Integrated ECG Platform for the Early Identification of
Electrocardiographic Markers Associated with Sudden Cardiac Death Risk in Young and Adult Populations
Project Description Sudden Cardiac Death (SCD) is one of the leading causes of cardiovascular mortality among apparently healthy young and adult individuals. In a significant proportion of cases, it represents the first clinical manifestation of an underlying electrical or structural heart disease that had previously remained undiagnosed. Several conditions associated with SCD, including Long QT Syndrome, Brugada Syndrome, Wolff-Parkinson-White (WPW) syndrome, ventricular arrhythmias, and advanced conduction disorders, exhibit electrocardiographic markers that can be identified during asymptomatic or minimally symptomatic stages. However, the effectiveness of ECG-based screening depends heavily on signal quality, standardized interpretation, availability of clinical history, and the ability to perform serial longitudinal comparisons. This project aims to evaluate the use of an integrated digital platform for advanced ECG management and analysis, with the objective of improving the early identification of individuals at risk of sudden cardiac death and supporting primary prevention strategies and structured follow-up programs. A dedicated artificial intelligence-based software solution will be used to screen ECG recordings, separating tracings without suspicious abnormalities (normal ECGs) from a second group of tracings that, although highly likely to be interpreted as normal by an experienced cardiologist, may contain subtle electrocardiographic features warranting further evaluation. Scientific Coordinators Prof. Maurizio Turiel Dr. Daniel Di Mattia In collaboration with: Objectives Primary Objective • To improve the early identification, through standard ECG and an integrated digital platform, of individuals presenting electrocardiographic markers associated with an increased risk of sudden cardiac death. Secondary Objectives • To assess the prevalence of clinically relevant electrocardiographic abnormalities, including pathological QTc prolongation, Type 1 Brugada ECG pattern, WPW syndrome, ventricular arrhythmias, and advanced conduction disturbances. • To analyze the temporal evolution of ECG parameters through serial ECG comparisons. • To generate and evaluate structured clinical datasets to support the implementation of future screening and prevention strategies. • To assess the feasibility of an extended monitoring model outside the hospital setting in selected patients. Expected Outcomes and Potential Applications The implementation of an integrated ECG platform is expected to provide: • Improved early identification of individuals presenting electrocardiographic markers associated with an increased risk of sudden cardiac death. • Reduced risk of underdiagnosing rare but clinically significant electrical disorders, such as Type 1 Brugada syndrome and markedly abnormal QTc prolongation. • Enhanced risk stratification through the availability of serial ECG recordings that can be compared longitudinally over time. • Assessment of the feasibility of large-scale screening programs and telecardiology- based follow-up models, including in young and apparently healthy populations. • Establishment of a structured ECG database suitable for epidemiological analyses and future clinical research studies. From a translational perspective, the project may serve as a replicable model for sudden cardiac death prevention programs, both in hospital-based and community healthcare settings, fully aligned with the mission and objectives of the Silvers Foundation. Target Population • Milan Jewish High School • Sports High Schools in the Milan metropolitan area • Sports Medicine Centers participating on a voluntary basis