Heart Failure Clinical Records for 299 Patients with 13 Features

299 patient records collected during a follow-up period, as described in the associated research paper. The dataset contains 13 clinical features for each patient, including age, ejection fraction, and serum creatinine. It was created by Chicco and Jurman for research published in BMC Medical Informatics and Decision Making in 2020.

Use Cases

• Predicting patient survival based on serum creatinine and ejection fraction levels.
• Analyzing the relationship between clinical features like anaemia or diabetes and mortality events.
• Building regression models to estimate platelet counts from other patient characteristics.
• Studying risk factors for heart failure mortality during a follow-up period.

Strengths

• Contains 299 real patient records with a documented follow-up period.
• Includes 13 well-defined clinical features, such as lab results and binary health indicators.
• Linked to a peer-reviewed research paper providing detailed context and validation.

Limitations

• Row count is known, but column-level details and sample data are unavailable for direct inspection.
• The temporal coverage of the data collection and the last update date are unknown.
• Data may reflect biases inherent to the specific patient cohort and collection methodology.

Provenance

Source

OpenML, originating from research by Chicco and Jurman.

Collection Method

Medical records collected during patient follow-up, as detailed in the associated paper.

Time Range

Follow-up period is included as a feature ('time'), but the overall collection date range is unspecified.

Freshness

Last updated date is unknown; freshness unverified.

Geography

Geographic coverage is not specified in the provided description.