RSNA RATIC: Abdominal Trauma CT Scans from 4,000+ Exams Across 14 Countries

More than 4,000 computed tomography (CT) exams of patients with and without abdominal traumatic injuries. The Radiological Society of North America collected this imaging data from 23 sites across 14 countries on six continents to create a ground truth dataset. It is intended to support the development of AI tools for expediting diagnosis in emergency settings.

Use Cases

• Train injury detection models based on CT scans showing damage to organs like the liver, spleen, kidneys, and bowel.
• Develop classification algorithms for abdominal trauma severity based on labeled imaging data.
• Benchmark AI-assisted diagnostic tools for emergency radiology based on a large, multi-site collection of trauma cases.
• Research imaging biomarkers for internal bleeding resulting from blunt force trauma mentioned in the description.

Strengths

• Contains more than 4,000 CT exams, providing substantial scale for model training.
• Includes a roughly equal number of cases with and without injury, which can aid in balanced classification tasks.
• Data was sourced from 23 sites in 14 countries across six continents, suggesting geographic diversity.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count and dataset size are unknown, which may limit suitability assessment.
• Last update date is unknown; freshness unverified.

Provenance

Source

Radiological Society of North America (RSNA)

Collection Method

Imaging data collected from 23 medical sites.

Time Range

null

Freshness

Last updated date is unknown.

Geography

Data sourced from 14 countries across six continents.