Back to evaluations
Public evaluation
MCED Guideline Query Task
Evaluates the factual fidelity and clinical precision of RAG generated diagnostic follow-up recommendations.
Evaluation type
task based
Challenge
LlamaIndex MCED Diagnostic Utility RAG with GPT-5 Pro
Difficulty
Intermediate
Rigor
Unspecified
Evaluation overview
How the linked challenge is judged: tasks, benchmarks, and criteria count.
Tasks
1
Benchmarks
0
Criteria
0
Task templates
Inputs and expected outputs.
Task 1
MCED Guideline Query Task
Query the RAG index with an MCED signal report and check compliance of recommended workup.
Input format
JSON with patient_age, mced_signal_tissue_of_origin, biomarker_level
Output format
JSON containing recommended_imaging, specialist_referral, and citations