Synthetic Cardiovascular Disease Dataset with 8,000 Patient Records

Martha Nyekanga's synthetic cardiovascular disease dataset contains 8,000 patient records and 12 variables. It was generated to simulate a diverse clinical population for testing machine learning models for CVD prediction. The dataset was last updated on April 30, 2026.

Use Cases

• Train binary classification models for CVD prediction based on demographic, clinical, and behavioral variables.
• Test algorithmic fairness across different ethnic and income groups as described in the dataset.
• Benchmark model performance on a synthetic clinical dataset with known prevalence and no missing values.

Strengths

• 8,000 patient records with no missing values.
• 12 variables covering demographic, clinical, and behavioral factors.
• CVD prevalence is explicitly set at approximately 19% (1,517 positive cases).
• Designed with proportionate representation across ethnic and income groups for fairness analysis.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Data is synthetic, which may limit real-world applicability.
• The dataset is small at 506.7 KB, indicating limited scope.

Provenance

Source

Martha Nyekanga via figshare.

Collection Method

Synthetically generated to simulate a diverse clinical population.

Time Range

null

Freshness

Last updated 2026-04-30 14:21:00; freshness should be verified.

Geography

null