This project demonstrates a Logistic Regression model built to predict heart failure events based on clinical features. The model is trained on a dataset containing various patient data points, which serve as predictors for the likelihood of a heart failure event. The final model achieves an accuracy of approximately 85%.
The dataset contains several features related to patient health, and each row represents data for a single patient. These features are used to train the model for predicting heart failure events.
- Age: Age of the patient (years).
- Anaemia: Whether the patient has anaemia (1 = Yes, 0 = No).
- Creatinine Phosphokinase (CPK): The level of the CPK enzyme in the blood (mcg/L).
- Ejection Fraction: Percentage of blood leaving the heart with each contraction.
- High Blood Pressure: Whether the patient has high blood pressure (1 = Yes, 0 = No).
- Platelets: Platelet count in the blood (kiloplatelets/mL).
- Serum Creatinine: Level of creatinine in the blood (mg/dL).
- Serum Sodium: Level of sodium in the blood (mEq/L).
- Sex: Gender of the patient (1 = Male, 0 = Female).
- Smoking: Whether the patient is a smoker (1 = Yes, 0 = No).
- Time: Follow-up period (days).
- DEATH_EVENT: Whether the patient died during the follow-up period (1 = Yes, 0 = No).
- DEATH_EVENT: The binary target variable representing the occurrence of a heart failure event (1 = death, 0 = no death).
The Logistic Regression algorithm was used to build this classification model. Logistic Regression is suitable for binary classification tasks like predicting whether a heart failure event occurs or not. The model uses the above features to predict the DEATH_EVENT outcome.
- Accuracy: The model achieves an accuracy of around 85%, indicating a strong performance in predicting heart failure events.