引導式測驗 · 中等 等級
AI 在遺傳變異判讀中的應用測驗
測試變異分類類別、證據類型、AI 的支持作用、ClinGen 規範、臨床解釋的限制。
問題 1 的 8
Which categories are included in the ACMG/AMP sequence-variant framework?
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。
- Which categories are included in the ACMG/AMP sequence-variant framework?
- Pathogenic, likely pathogenic, uncertain significance, likely benign, and benign
- High, medium, and low risk only
- Diagnostic, prognostic, and therapeutic
- Positive, negative, and inconclusive test result only
- What does a computational prediction contribute to variant interpretation?
- Proof that a variant causes disease
- A substitute for phenotype and inheritance review
- One evidence type to assess alongside other evidence and criteria
- A standalone clinical classification
- How should a variant of uncertain significance be described?
- It is equivalent to a pathogenic variant
- It proves the person has the associated disorder
- It can always guide treatment as a positive result
- Its clinical significance is unresolved; it is not a confirmed disease-causing finding
- Why check the original paper after an AI tool retrieves it?
- A model can determine whether the evidence is clinically valid from its title
- The paper may involve a different variant, assay, disease, or inheritance context
- The retrieval score already confirms every claim
- The abstract guarantees the clinical classification
- How does a ClinGen expert-panel specification guide variant curation?
- It converts a model ranking directly into a diagnosis
- It removes the need to review the patient phenotype
- It refines how variant criteria are applied for a defined gene or disorder
- It replaces ACMG/AMP evidence categories across all diseases
- Which evidence types may contribute to a classification?
- Population, computational, functional, segregation, and clinical evidence
- Only a language model’s confidence score
- Only the variant’s position in the genome
- Only a single family member’s reported symptom
- What should be preserved for a reproducible curation decision?
- A screenshot of the classification with no provenance
- Source evidence, applied criteria, date, reviewer, and conflicts
- Only the model’s final ranking
- The patient’s name in a public prompt
- What can an AI variant-prioritization score establish by itself?
- That a variant is pathogenic
- That a family has a specific disease
- That treatment should begin
- It can rank candidates for review but does not establish a clinical classification