引導式測驗 · 中等 等級

人工智慧在腫瘤治療計畫測驗的應用

測試腫瘤學決策支援、Watson 一致性與結果、腫瘤委員會審查以及臨床醫生對建議的獨立評估。

問題 1 的 8

What does an oncology decision-support system contribute to treatment planning?

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 oncology decision-support system contribute to treatment planning?
    • It organizes possible options for clinicians to review
    • It proves which regimen gives each patient the best survival
    • It prescribes the highest-ranked regimen automatically
    • It replaces tumor-board review with a model score
  2. What did the cited Watson for Oncology breast-cancer study measure?
    • A randomized difference in survival between AI and usual care
    • Diagnostic accuracy across all cancer types and hospitals
    • The number of patients cured by use of the system
    • Agreement with a tumor board under its stated concordance definition
  3. Why does a concordance percentage not prove a treatment system improves outcomes?
    • The score proves the system caused better outcomes
    • Agreement is a process measure, not evidence of patient benefit
    • Concordance measures patient survival directly
    • A match with one board guarantees the best choice for everyone
  4. Which patients were less likely to have concordant recommendations in the cited Watson breast-cancer study?
    • Patients with receptor status recorded in the chart
    • Patients treated at a hospital with a tumor board
    • Patients with stage I or IV disease and older patients
    • Patients whose case included prior treatment
  5. What should a tumor board do with an AI-suggested regimen?
    • Compare it with the record, current evidence and local availability
    • Adopt the first-ranked option without discussing alternatives
    • Ignore patient preferences if guidelines are cited
    • Use the same plan for every patient with that cancer type
  6. What does FDA guidance emphasize for clinician-facing decision support?
    • The software should hide its inputs to avoid bias
    • Only the vendor needs to understand how the options were ranked
    • A fast answer should replace a time-consuming tumor board
    • Clinicians should be able to review the basis for a recommendation
  7. Why can an option suggested in one hospital be unsuitable in another?
    • Every hospital follows the same formulary
    • Drug availability and treatment practices can differ
    • Clinical guidelines never change
    • A model automatically knows every local constraint
  8. What remains necessary after an AI system lists treatment choices?
    • The system issues a prescription without clinician review
    • The software vendor approves the final plan
    • A qualified care team and patient discuss evidence, risks and goals
    • The patient skips discussion of side effects