Canonical text (the exact bytes fingerprinted)

TRUVACE RECORD VERSION
record: TRV-2026-0937
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-08-31T06:05:54.561618Z
status: published
lens: g_space
sector: health
headline: Ai-based multimodal analysis of ECG and clinical data for evaluation for competitive sports participation: The VALETUDO trial
dek: Aims Pre-participation cardiovascular screening (PPS) is essential for preventing SCD in athletes, yet ECG interpretation requires expertise and remains resource-intensive. We aimed to evaluate the feasibility and diagnostic performance of a deep learning (DL) model for analysis of clinical data and resting 12‑lead ECG obtained during routine PPS in competitive athletes. Methods In this prospective single center observational study, competitive athletes aged 18 to 60 years and undergoing routine PPS were enrolle…
gain_title: A multimodal deep learning model analyzing resting 12-lead ECG and clinical data was feasible for pre-participation cardiovascular screening and achieved moderate discrimination for fitness for competitive sports.
problem_title: (none)
trace_subject: (none)
gain_reading: A multimodal deep learning model analyzing resting 12-lead ECG and clinical data was feasible for pre-participation cardiovascular screening and achieved moderate discrimination for fitness for competitive sports.
gain_evidence: Automated DL-based analysis of 12lead ECG during PPS is feasible and showed encouraging diagnostic performance in competitive athletes.
problem_reading: (none)
problem_evidence: (none)
quick_read: In a prospective single-center study of 526 competitive athletes undergoing routine pre-participation cardiovascular screening, researchers tested a multimodal deep learning model that combined resting 12-lead ECG with clinical variables to predict clinical fitness classification. Using stratified 10-fold cross-validation, the model achieved test accuracy of 0.64 b1 0.08 and AUC 0.72 (0.66-0.78).

The result suggests automated ECG interpretation is feasible as an adjunct to expert physician assessment in sports cardiology screening, potentially reducing resource intensity. However, moderate discrimination, low NPV of 0.48 b1 0.12, and lack of external validation leave uncertainty about generalizability and clinical safety before deployment.
limitation: Single-center design with 526 athletes and no external validation, requiring wider experience before clinical use.
tag: Evidence-backed gain
key_points: Prospective single-center observational study enrolled 526 competitive athletes aged 18 to 60 years undergoing routine PPS. | PPS included medical history, physical examination, resting and exercise ECG, with athletes classified as fit or not fit by clinical evaluation. | Model performance was assessed using stratified 10-fold cross-validation against PPS clinical classification. | Test setting performance was accuracy 0.64 b1 0.08, sensitivity 0.68 b1 0.15, specificity 0.61 b1 0.17, AUC 0.72 (0.66-0.78).
rundown: The VALETUDO trial enrolled 526 competitive athletes, 72% male with median age 27 years (IQR: 20-41), of whom 166 (32%) had a negative PPS result. Resting ECG and clinical variables were analyzed using a multimodal DL architecture and compared to physician-determined fitness classification.

In 10-fold cross-validation the model reached training accuracy 0.70 b1 0.06 and AUC 0.79 (0.74-0.83), dropping to test accuracy 0.64 b1 0.08 and AUC 0.72 (0.66-0.78). Authors concluded AI-assisted interpretation may be a useful adjunct to physician assessment for cardiovascular risk stratification.
sources:
- peer_reviewed | International Journal of Cardiology | https://doi.org/10.1016/j.ijcard.2026.134744 | 2026-08-29
prev: 0000000000000000000000000000000000000000000000000000000000000000

sha256

50829e4a880ab678444458e27d7e021b49b90e9c698622eeadd2321616bdd984

previous

0000000000000000000000000000000000000000000000000000000000000000