Dark Mode
Heart Failure Mortality Prediction Dataset
Public Health & Epidemiology
Tags and Keywords
Trusted By
"No reviews yet"
Free
About
This dataset focuses on predicting mortality in patients with heart failure, a critical application given that cardiovascular diseases (CVDs) are the leading global cause of death. It contains 12 clinical features that are instrumental in developing machine learning models to identify patients at high risk of death due to heart failure. The data can aid in early detection and management strategies, potentially preventing a significant number of yearly fatalities associated with CVDs.
Columns
- age: Age of the patient.
- anaemia: A boolean indicator for the presence of anaemia (decrease of red blood cells or haemoglobin).
- creatinine_phosphokinase: Level of the CPK enzyme in the blood, measured in mcg/L.
- diabetes: A boolean indicator for whether the patient has diabetes.
- ejection_fraction: Percentage of blood leaving the heart at each contraction.
- high_blood_pressure: A boolean indicator for the presence of hypertension (high blood pressure).
- platelets: Platelets in the blood, measured in kiloplatelets/mL.
- serum_creatinine: Level of serum creatinine in the blood, measured in mg/dL.
- serum_sodium: Level of serum sodium in the blood, measured in mEq/L.
- sex: A binary indicator for gender (woman or man).
- smoking: A boolean indicator for whether the patient smokes.
- time: The follow-up period for the patient, measured in days.
- DEATH_EVENT: The target variable, a boolean indicator for whether the patient deceased during the follow-up period.
Distribution
The dataset is provided as a CSV file named
heart_failure_clinical_records_dataset.csv. It has a file size of 12.24 kB and contains 299 records (rows). All 13 columns are fully populated, with no missing values.Usage
This dataset is ideal for building and evaluating machine learning models aimed at predicting mortality caused by heart failure. It can be used by researchers and data scientists to:
- Develop predictive analytics for patient risk stratification.
- Explore the significance of various clinical features in heart failure outcomes.
- Create tools for early detection and management of cardiovascular risks.
- Contribute to healthcare informatics by advancing AI applications in cardiology.
Coverage
The dataset includes clinical records with features such as age and sex (binary for woman or man), providing some demographic scope. The
time column specifies a follow-up period in days, indicating the duration of observation for each patient up to 285 days. The dataset does not explicitly state geographic coverage or specific data collection dates.License
CC BY 4.0
Who Can Use It
This dataset is primarily intended for data scientists, machine learning engineers, and researchers in the fields of healthcare, medical informatics, and public health. It is particularly valuable for those looking to:
- Build and benchmark mortality prediction models.
- Analyse clinical features associated with heart failure.
- Innovate in preventative healthcare using data-driven insights.
Dataset Name Suggestions
- Heart Failure Mortality Prediction Dataset
- Cardiovascular Death Prediction Data
- Clinical Heart Failure Outcomes
- Patient Heart Failure Survival Records
- CVD Mortality Risk Dataset
Attributes
Original Data Source: Heart Failure Mortality Prediction Dataset
QUALITY