引導式測驗 · 中等 等級
AI 在肺癌篩檢 CT 測驗
測試 AI 結節支持、篩檢與診斷、FDA 預期用途、假陽性風險、放射科醫生審查和後續要求。
問題 1 的 8
What output is expected from the FDA-authorized lung CT CAD tool described here?
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。
- What output is expected from the FDA-authorized lung CT CAD tool described here?
- Decide who is eligible for screening
- Replace follow-up imaging after a positive result
- Highlight potential nodules for radiologist review
- Diagnose lung cancer independently from the image
- How does screening CT differ from a diagnostic CT?
- There is no difference in purpose
- Screening is used for eligible people without symptoms; diagnostic imaging investigates a clinical concern
- Screening is only performed after a nodule is confirmed
- Diagnostic CT is always AI-read and screening is not
- What can a positive AI nodule mark mean?
- A confirmed cancer diagnosis
- Proof that the person must have surgery
- Evidence that the nodule is malignant
- A candidate finding for radiologist review that may lead to follow-up
- Why can an AI nodule tool produce false positives?
- Normal structures or benign nodules may resemble the patterns it flags
- The model reads the patient’s family history directly
- The scanner automatically labels every nodule as cancer
- Radiologists never compare prior images
- What should a radiologist do with a model-generated nodule mark?
- Change the screening eligibility rules
- Review it in the source CT with clinical context and prior images
- Accept it automatically as a diagnosis
- Ignore image quality and patient history
- Why validate AI performance on local scanners and populations?
- Image protocols and patient populations can affect model performance
- FDA authorization guarantees identical performance everywhere
- Local validation changes the medical diagnosis
- A different scanner cannot alter image quality
- Which screening harm should patients discuss with a clinician?
- The guarantee that no cancer can be missed
- An automatic treatment from the AI tool
- The removal of all incidental findings
- False positives and follow-up tests after an abnormal finding
- What does an AI result alone establish?
- That cancer is present or absent with certainty
- That follow-up is unnecessary
- Only a software output that must be interpreted in the screening pathway
- That the patient meets screening criteria