Hospital EHR Data: Synthetic Records for 1,171 Patients

1,171 synthetic patient records across 15 tables provide privacy-safe electronic health data. The dataset is designed for machine learning applications in clinical settings. Its origin and specific collection details are not provided.

Use Cases

• Train predictive models for patient outcomes based on synthetic EHR features.
• Benchmark clinical data anonymization techniques using the privacy-safe synthetic data.
• Develop and test healthcare data pipelines with a multi-table relational structure.
• Simulate clinical decision support systems using realistic but synthetic patient records.

Strengths

• Contains 1,171 patient records.
• Structured across 15 relational tables.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count per table is unknown, which may limit suitability assessment.
• Data may reflect bias inherent to its unspecified synthetic generation method.

Provenance

Collection Method

Synthetic generation.