Synthetic EHR Dataset with Text and Vital Signs

A synthetic electronic health record dataset integrating text notes and time-series vital sign data. The dataset is designed for healthcare predictive research, specifically HPR. It was created by an unknown author and published on Kaggle, with no information on its size or last update.

Use Cases

• Train models to predict clinical outcomes from integrated text notes and time-series vital signs.
• Develop and benchmark multimodal fusion architectures using synthetic EHR text and numerical time-series data.
• Simulate patient trajectories by modeling the relationship between documented text events and recorded vital sign trends.
• Create synthetic training data for downstream tasks like risk stratification or phenotyping using combined clinical modalities.

Strengths

• Integrates multiple data modalities (text and time-series) within a clinical context.
• Synthetic nature allows for use without privacy restrictions associated with real patient data.

Limitations

• Unknown row count prevents assessment of dataset scale and statistical power.
• Synthetic data may not fully capture the complexity and noise of real-world clinical records.
• Lack of column details limits understanding of specific features and data structure.

Provenance

Source

Kaggle

Collection Method

Synthetically generated.