Evaluate whether models can detect conflicts between patient information and proposed care levels in structured general practice triage cases.

Test whether models can verify pediatric vaccination plans against age requirements, dose intervals, and contraindications using structured cases and adjudicated results.

Evaluate community medicine constraint reasoning with structured cases covering screening eligibility, care pathways, and follow-up deadlines.

Evaluate schedule-checking capabilities with adolescent cases paired with versioned vaccination constraints and reference outcomes.

Evaluate whether models can detect contradictions between clinical findings and diagnoses in retinal, macular, and vitreous records.

An image recognition dataset for medical sleep quality monitoring instruments, aimed at improving instrument detection accuracy.

Evaluate whether models can reason across eligibility constraints for pelvic floor treatments and procedures using structured cases and rule-based checks.

The Medical Teaching Image Dataset is a high-quality image resource to assist medical teaching.

Evaluate how well models judge eligibility and timing for general practice screening and preventive care against explicit patient-specific constraints.

Evaluate whether models can detect medication constraint violations in surgical critical care cases involving dosage, allergies, organ function, and drug interactions.

Evaluate a model's ability to detect conflicts in structured rhinology cases with reference consistency labels and contradiction annotations.

The ICU Equipment Interface Image Recognition Dataset is used to enhance the accuracy of medical equipment interface recognition.

Evaluate how well models detect omissions and constraint conflicts in proposed dispositions using case risk information and triage rules.

Evaluate how well models verify evidence against pediatric ADHD criteria, one diagnostic constraint at a time.

Evaluate how well models identify ordering errors, dependencies, and operational conflicts in neurosurgical workflows using reference violation labels.

The Clinic Electronic Medical Record Q&A Dataset focuses on image data mining in the healthcare field.

Compare case evidence with diagnostic criteria to evaluate a model's ability to find missing conditions and contradictions.

Evaluate how well a model detects care pathway conflicts in pediatric febrile neutropenia by checking case details and proposed management against pathway rules.

A high-quality dataset to enhance the accuracy of drug image recognition in intelligent pharmacy systems.

Follow the full path from identifying a flaw in an acute migraine treatment plan to agent self-correction, and train more reliable medical agents.