Dark Mode
Heart Failure Biomarker Dataset
Patient Health Records & Digital Health
Tags and Keywords
Trusted By
"No reviews yet"
Free
About
This dataset provides clinical records for patients diagnosed with heart failure, primarily intended for predicting patient survival outcomes. It encompasses a collection of medical records from 5000 individuals during their follow-up period, each characterised by 13 distinct clinical features. The primary objective is to facilitate the development of models that can forecast whether a patient will survive or succumb during their follow-up. This resource offers valuable insight for medical research and predictive analytics in cardiovascular health.
Columns
- age: The patient's age in years.
- anaemia: A boolean indicator (0 or 1) denoting a decrease in red blood cells or haemoglobin.
- creatinine phosphokinase (CPK): The level of the CPK enzyme in the blood, measured in micrograms per litre (mcg/L).
- diabetes: A boolean indicator (0 or 1) if the patient has diabetes.
- ejection fraction: The percentage of blood leaving the heart with each contraction.
- high blood pressure: A boolean indicator (0 or 1) if the patient has hypertension.
- platelets: The count of platelets in the blood, measured in kiloplatelets per millilitre (kiloplatelets/mL).
- sex: A binary indicator (0 for female, 1 for male).
- serum creatinine: The level of serum creatinine in the blood, measured in milligrams per decilitre (mg/dL).
- serum sodium: The level of serum sodium in the blood, measured in milliequivalents per litre (mEq/L).
- smoking: A boolean indicator (0 or 1) if the patient smokes.
- time: The duration of the follow-up period in days.
- DEATH_EVENT: A boolean indicator (0 or 1) signifying if the patient died during the follow-up period.
Distribution
The dataset is typically provided as a CSV file, named
heart_failure_clinical_records.csv. It has a file size of approximately 223.27 kB and contains records for 5000 patients. The structure is tabular, consisting of 13 columns. Each column has 5000 valid entries with no mismatched or missing values.Usage
This dataset is ideal for:
- Developing machine learning models for classification, specifically to predict patient mortality due to heart failure.
- Data visualisation to explore relationships between clinical features and patient outcomes.
- Conducting medical research focused on risk factors and prognosis in heart conditions.
- Training and evaluating predictive algorithms in a healthcare context.
Coverage
The dataset covers clinical records of patients who experienced heart failure.
- Geographic Scope: Not explicitly stated, but implies a medical clinic or hospital setting where these records were collected.
- Time Range: The
timecolumn represents the follow-up period, ranging from 4 to 285 days for different patients. The dataset itself was cited in 2020. - Demographic Scope: Includes patients of varying ages (from 40 to 95 years) and both sexes (male and female). It captures various health statuses, including the presence of anaemia, diabetes, and high blood pressure, as well as smoking habits.
License
CC0: Public Domain
Who Can Use It
- Data Scientists and Machine Learning Engineers for building and testing predictive models.
- Medical Researchers and Healthcare Analysts to understand clinical factors influencing heart failure outcomes.
- Students and Academics learning about classification algorithms and healthcare data analysis.
- Public Health Professionals interested in cardiovascular health trends and patient risk assessment.
Dataset Name Suggestions
- Heart Failure Patient Outcomes
- Clinical Cardiac Mortality Prediction
- Cardiovascular Survival Records
- Heart Failure Biomarker Dataset
- Patient Heart Health Predictor
Attributes
Original Data Source: Heart Failure Biomarker Dataset
QUALITY