TREC 2022 Clinical Trials Matching Dataset

The TREC 2022 Clinical Trials Dataset focuses on the clinical trials matching problem. It contains patient health record summaries and associated clinical trial suitability assessments, curated by the National Institute of Standards and Technology for the Text Retrieval Conference track.

Use Cases

• Train information retrieval models to match free-text patient summaries to suitable clinical trials using the TREC QRELS relevance judgments.
• Benchmark natural language processing systems on the clinical trials matching task defined by the TREC track.
• Analyze the structure and content of patient health record summaries provided in XML or HTML formats for retrieval research.

Strengths

• Dataset is curated by the National Institute of Standards and Technology (NIST) for a formal TREC evaluation track.
• Data is designed for a specific, well-defined research problem: clinical trials matching from patient summaries.
• Includes TREC QRELS files, which provide standardized relevance judgments for evaluation.

Limitations

• The dataset size, row count, and specific column structure are unknown, limiting pre-use assessment.
• Data is from a 2022 evaluation track and may not reflect the most current clinical trial or patient record formats.
• The 'other-license-specified' license requires careful review for usage rights and restrictions.

Provenance

Source

National Institute of Standards and Technology (NIST)

Collection Method

Created for the TREC (Text Retrieval Conference) 2022 Clinical Trials track.

Time Range

2022

Freshness

Last updated in September 2024, though the core data is from the 2022 TREC track.