引導式測驗 · 中等 等級

醫院感染預防中的 AI 測驗

測試 NHSN 標準、監測與診斷、警報驗證、報告可比性和感染預防審查。

相關引導路徑醫院感染預防中的 AI
問題 1 的 8

What does an AI HAI alert establish?

預覽每個問題8 問題

當您參加測驗時,就會出現答案和解釋。

  1. What does an AI HAI alert establish?
    • Proof the event meets a surveillance definition
    • A case for review against surveillance criteria and the patient record
    • A confirmed patient-specific diagnosis
    • A mandatory antimicrobial order
  2. What role do CDC NHSN definitions play?
    • Optional labels that replace clinical documentation
    • Rules that determine antibiotic choice automatically
    • Definitions that replace all bedside assessment
    • Criteria for standardized infection surveillance and reporting
  3. Which information gap can make surveillance signals insufficient to distinguish colonization from infection?
    • Lab and clinical signals may be incomplete or lack context
    • Surveillance definitions are optional for standardized reporting
    • Every positive culture confirms symptomatic infection regardless of context
    • A surveillance criterion may be confused with a patient-specific diagnosis
  4. How should a hospital validate an infection-surveillance model?
    • Compare alerts with current criteria and chart review by infection type and unit
    • Compare alerts only with prior model predictions and aggregate counts
    • Use one accuracy threshold across all infections without chart review
    • Ask the model vendor to certify the result without local validation
  5. Why document changes to case identification?
    • To remove all historical results when a model changes
    • Because they can affect comparability of infection rates over time
    • To ensure that infection rates always decline after deployment
    • Because case definitions are private model parameters
  6. Which monitoring set covers both detection quality and workflow for deployed infection flags?
    • Retrospective sensitivity and specificity only, without timing or reviewer measures
    • Alert volume and acceptance, without checking missed cases or delay
    • False alerts, missed cases, timeliness, and reviewer agreement
    • Discrimination and calibration only, with no live workflow measures
  7. Should a surveillance flag alone trigger treatment?
    • No; treatment requires clinical assessment and evidence
    • Yes, if a score and positive result replace patient-specific assessment
    • Yes, once a surveillance definition matches, because treatment follows automatically
    • No; surveillance results should be ignored during clinical assessment
  8. What may AI support in infection prevention?
    • Replacing standard surveillance criteria
    • Declaring every infection automatically
    • Removing the need for prevention staff
    • Prioritizing record review or identifying patterns for human assessment