Alessandro Biffi

Background: Artificial intelligence (AI) has shown promise in the interpretation of electrocardiograms (ECGs) using signal-based deep learning models. In parallel, large language models (LLMs) have gained increasing visibility in clinical practice, including exploratory applications in ECG analysis. Whether a general-purpose LLM can meaningfully discriminate cardiovascular disease from athlete ECGs during PPS remains unknown. We aimed to evaluate the diagnostic performance of a general-purpose LLM for this task. Methods: In this multicentre diagnostic accuracy study, we evaluated a commercially available LLM (ChatGPT, version 5) in 2950 competitive athletes undergoing PPS. All athletes underwent resting 12-lead ECG, with second- and third-line investigations performed when clinically indicated. The reference outcome was confirmed cardiovascular disease after full diagnostic work-up (n = 450, 15.3%). For each ECG, the LLM generated a numeric score (0–100) representing the inferred likelihood of underlying disease using a standardized prompt and without task-specific fine-tuning. Discriminative performance was assessed using receiver operating characteristic (ROC) analysis. Misclassification patterns were analysed according to International ECG Criteria. Results: GPT-derived scores demonstrated a marked floor effect, with a median value of 0 (IQR 0–2) in both diseased and non-diseased athletes and substantial overlap between groups. The area under the ROC curve was 0.52 (95% CI 0.49–0.55), indicating performance close to random classification. At the Youden-derived threshold, 79% of athletes with confirmed disease were incorrectly classified as negative. False-negative cases were predominantly characterized by borderline ECG patterns (82%), and a substantial number of red-flag ECG abnormalities were also missed. Conclusions: In this PPS cohort, a general-purpose LLM used in a naïve configuration showed no clinically meaningful ability to discriminate between cardiovascular disease and athlete ECGs. Without task-specific training or domain adaptation, such models should not be used for diagnostic triage in athlete screening.

Introduction Cardiovascular (CV) diseases are the leading cause of mortality worldwide, and preventive strategies are increasingly focusing on early detection and management of risk factors. Corporate Wellness Programs (CWPs) in workplaces play a crucial role in the primary prevention of chronic non-communicable diseases, especially for early CV modifiable risk detection and management in apparently healthy individuals. Aim This study explores the prevalence and age-related distribution of CV modifiable risk factors in a large cohort of Ferrari car manufacturer employees evaluated through a CWP. Methods A cross-sectional analysis was conducted among 1992 employees participating for the first time in the “Ferrari Formula Benessere” CWP in Maranello, Italy, throughout 2023. The program included a comprehensive medical evaluation, personal and family history, anthropometric measurements, blood testing, and blood pressure evaluation. Data on demographic characteristics, health behaviors, and CV modifiable risk factors were collected and analyzed. Results Most participants were males (82%), with a mean age of 40 ± 9 years. Findings revealed a notable prevalence of CV modifiable risk factors: 17.3% were active smokers, 43.2% reported physical inactivity, 34.8% were classified as overweight (body mass index ≥ 25), and 7.7% as obese. Early screening identified a significant proportion of employees with elevated CV risk according to the SCORE2 algorithm: 39% were categorized as having a high CV risk and 2% at a very high CV risk. In addition, 6.4% had impaired fasting glycemia (100–125 mg/dL), 3.8% had dyslipidemia and 6.8% had high blood pressure (systolic blood pressure 140–159 mmHg): almost none of them (98%) were on treatment. Conclusions The “Ferrari Formula Benessere” identified a wide range of CV modifiable risk factors in a cohort of apparently healthy employees, not routinely screened, highlighting the essential role of CWPs in the early detection and management of CV health risks. These findings underscore the importance of integrating regular health assessments into corporate wellness strategies to mitigate CV risk and promote overall employee well-being.

Background. Poor posture is a common musculoskeletal concern in children and adolescents and may lead to spinal discomfort and long-term structural issues. While excess weight has been linked to altered sagittal alignment, the impact of sports participation on spinal posture remains unclear. This study aimed to investigate the associations between weight status, sports participation, and sagittal spinal alignment in a pediatric population. Methods. This cross-sectional study was conducted within the “Ferrari Formula Benessere” corporate wellness program and included 698 children aged 5 to 16 years. Sagittal Index (SI) was measured using a standardized plumb line technique. Body mass index (BMI) was calculated and classified according to the WHO growth standards. Sports participation was self-reported and grouped into five categories: sedentary, skill-based, power-based, mixed, and endurance sports. Results. Age was the only significant independent predictor of the SI (β = 2.45, p < 0.001), with older children exhibiting higher SI values. Although a weak correlation was observed between BMI and SI (Spearman’s r = 0.24, p < 0.001), BMI was not a significant predictor when controlling for age. No significant differences in the SI were found between active and non-active children. Among sport disciplines, the SI was lowest in power-based sports (56.7 ± 22.3 mm) and higher in endurance (62.7 ± 24.4 mm), mixed (64.5 ± 23.2 mm), skill-based (61.1 ± 22.0 mm), and non-sport (64.2 ± 24.0 mm) groups, although these differences did not reach statistical significance (ANOVA p = 0.224). Conclusions. Age appears to be the primary factor associated with sagittal spinal alignment in children, while BMI and general sports participation showed no independent effect. Although some differences emerged between sport types, these findings were not statistically significant and should be interpreted with caution. These findings underscore the need for sport-specific, longitudinal research using objective posture assessment methods.

Box 1 High-risk clinical and ventricular arrhythmia features in competitive athletes

Purpose: The “FIDE Project” (Fitness Implantable DEvice) was organized by the Institute of Sports Medicine and Science and the World Society of Arrhythmias with the aim of demonstrating the usefulness of exercise training in improving functional capacity in patients with implantable cardiac devices. Materials and Methods: Thirty sedentary patients were selected for the project (25 males and 5 females), with a mean age of 73 ± 5 years (range 44–94 years). Twenty-five were implanted with a Pacemaker (PM) and five with an Implantable Cardioverter Defibrillator (ICD). Atrial fibrillation/atrial flutter was present in ten (34%) patients, post-ischemic dilated cardiomyopathy in five (17.2%), sick sinus syndrome in six (20,7%), complete atrium-ventricular block in six (20.7%), hypertrophic cardiomyopathy in one (3.4%) and recurrent syncope in one (3.4%). The baseline assessment comprised cardiovascular examination, resting and stress ECG, cardiopulmonary exercise testing (V ̇O2peak), strength assessment of different muscle groups, and a flexibility test. The same measurements were repeated after 15–20 consecutive training sessions, over a 2-month period. The exercise prescription was set to 70–80% of HRR (Heart rate reserve) and to 50–70% of 1RM (1-repetition maximum, muscular force). The training protocol consisted of two training sessions per week performed in our institute, 90 min for each (warm-up, aerobic phase, strength phase and stretching) and one or more at home autonomously. Results: The cardiopulmonary testing after the training period documents a significant improvement in V ̇O2peak (15 ± 4 mL/kg/min vs. 17 ± 4; p = 0.001) and in work load (87 ± 30 watts vs. 108 ± 37; p = 0.001). Additionally, strength capacity significantly increased after the cardiac rehabilitation program, (quadriceps: 21 ± 18 kg vs. 29 ± 16 kg, p = 0.00003). Flexibility tests show a positive trend, but without statistical significance (sit-and-reach test: −19 ± 11 cm vs. −15 ± 11.7 cm; back-scratch test: −19 ± 11.6 cm vs. −15 ± 10 cm; lateral flexibility right −44 ± 1.4 cm vs. −43 ± 9.5 cm; left −43 ± 5 vs. −45 ± 8.7 cm). Conclusion: A brief period of combined aerobic, strength and flexibility exercise training (FIDE project) proved to be effective and safe in improving functional capacity in patients with cardiac implantable devices.

In Italy the existence of a law on health protection of competitive sports since 1982 has favored the creation and the revision of these cardiological guidelines (called COCIS), which have reached their fourth edition (1989–2009). The present article is the second English version, which has summarized the larger version in Italian. The experience of the experts consulted in the course of these past 20 years has facilitated the application and the compatibility of issues related to clinical cardiology to the sports medicine field. Such prolonged experience has allowed the clinical cardiologist to acquire knowledge of the applied physiology of exercise and, on the other hand, has improved the ability of sports physicians in cardiological diagnostics. All this work has produced these guidelines related to the judgment of eligibility for competitive sports in the individual clinical situations and in the different cardiovascular abnormalities and/or heart disease. Numerous arguments are debated, such as interpretation of the athlete's ECG, the utility of a preparticipation screening, arrhythmias, congenital heart disease, cardiomyopathies, arterial hypertension, ischemic heart disease and other particular issues.

In Italy the existence of a law on health protection of competitive sports since 1982 has favored the creation and the revision of these cardiological guidelines (called COCIS), which have reached their fourth edition (1989-2009). The present article is the second English version, which has summarized the larger version in Italian. The experience of the experts consulted in the course of these past 20 years has facilitated the application and the compatibility of issues related to clinical cardiology to the sports medicine field. Such prolonged experience has allowed the clinical cardiologist to acquire knowledge of the applied physiology of exercise and, on the other hand, has improved the ability of sports physicians in cardiological diagnostics. All this work has produced these guidelines related to the judgment of eligibility for competitive sports in the individual clinical situations and in the different cardiovascular abnormalities and/or heart disease. Numerous arguments are debated, such as interpretation of the athlete's ECG, the utility of a preparticipation screening, arrhythmias, congenital heart disease, cardiomyopathies, arterial hypertension, ischemic heart disease and other particular issues.