Sztuczna inteligencja w quizie dotyczącym przewidywania kontuzji sportowca

Odpowiedzi i wyjaśnienia pojawiają się w miarę rozwiązywania quizu.

1. What does the acute-to-chronic workload ratio summarize?
   • A recent workload relative to a longer-term workload summary
   • A player’s injury severity relative to salary
   • A team’s travel time relative to match time
   • Heart-rate variability relative to sleep duration
2. Why can injury models be difficult to evaluate when the outcome is uncommon?
   • Every athlete has the same outcome
   • A model can appear accurate while missing many rare injury events
   • Rare events eliminate the need for validation
   • A high accuracy score guarantees useful prediction
3. Which measurements did the Australian football workload study use?
   • Ticket scans and social-media posts
   • Only heart-rate variability
   • GPS, accelerometer, and player-perceived exertion data
   • Only video of match officials
4. What does an injury-forecasting model estimate?
   • A guaranteed date of injury
   • A clinical diagnosis without examination
   • A certain training prescription
   • A probability or risk estimate for a defined outcome and period
5. What did Carey et al. report for non-contact injury prediction on the later season?
   • Performance was only marginally better than chance, with AUC below 0.65
   • Every outcome had perfect accuracy
   • AUC was always 1.0
   • The study did not test a later season
6. What validation gap did the 2026 systematic review report for all included studies?
   • All used a separate country for testing
   • None used external validation
   • All models were tested prospectively in clinics
   • No study reported any outcome metrics
7. Which split did the Australian football study use to test its injury forecasts?
   • It mixed all three seasons randomly for every result
   • It trained only on injury labels from the third season
   • It trained using the first two seasons and evaluated predictions on the third
   • It evaluated only on the training rows
8. What do sensitivity and specificity measure in an injury model?
   • They both measure only training speed
   • Sensitivity is cost and specificity is model size
   • They are alternative names for data leakage
   • Sensitivity concerns detecting injured cases; specificity concerns correctly identifying non-injured cases