Using SHAP and LIME for AI Explainability
A healthcare AI system that diagnoses medical images must provide explanations for its predictions to comply with regulatory requirements. Which technique should the team implement?
⚠ Common exam trap
It's easy for candidates to assume complex models are inherently better for compliance, but the exam tests the understanding that interpretability techniques are required to bridge the gap between high-performance black-box models and regulatory transparency.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Apply model interpretability methods such as SHAP or LIME.
SHAP (SHapley Additive exPlanations) and LIME (Local Interpretable Model-agnostic Explanations) are established model interpretability techniques that provide per-prediction explanations, which are essential for regulatory compliance in healthcare AI. These methods generate feature attribution scores or local surrogate models to explain why a specific diagnosis was made, meeting transparency requirements without sacrificing model performance.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Reduce the model's accuracy to make it simpler.
Why it's wrong here
Reducing accuracy does not generate explanations; it merely degrades diagnostic performance, so the regulatory requirement for interpretable predictions remains unmet. The temptation lies in conflating simplicity with transparency, since inherently interpretable models such as decision trees or linear regression are appropriate when explainability is required by design and accuracy loss is acceptable.
- ✗
Only deploy rule-based systems.
Why it's wrong here
Rule-based systems cannot learn from image data, so they cannot perform the diagnostic task at all; the requirement is to explain an existing model's predictions, not replace it. It is tempting because rule-based logic is fully transparent by construction, and it would be the correct choice where the domain can be fully encoded as explicit rules.
- ✓
Apply model interpretability methods such as SHAP or LIME.
Why this is correct
SHAP and LIME are post-hoc interpretability techniques that attribute a model's output to individual input features, generating per-prediction explanations. This satisfies the regulatory requirement for transparent, auditable diagnostic reasoning without retraining, unlike inherently opaque deep networks or purely performance-focused methods.
- ✗
Use a more complex deep learning model.
Why it's wrong here
Increasing model complexity reduces interpretability, so the system would produce predictions with even less explanation available. It is tempting because deeper networks often raise diagnostic accuracy on medical imaging, and complexity would be the right choice when accuracy is the sole objective and explainability is not required.
About these practice questions
One of 962 original AI0-001 practice questions on Courseiva, each with a full explanation and wrong-answer analysis — not exam dumps or protected exam content. Learn why practice questions differ from exam dumps →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This AI0-001 practice question is part of Courseiva's free CompTIA certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the AI0-001 exam.