A machine learning project that predicts the likelihood of death events in heart failure patients using clinical data and classification models.
This project focuses on predicting heart failure outcomes using clinical data and machine learning techniques. The goal is to assist in early detection and decision-making by analyzing patient health records.
Heart disease is one of the leading causes of death worldwide. By leveraging machine learning, this project aims to provide a data-driven approach to predict patient survival and support healthcare insights.
The dataset contains medical records of patients with heart failure and includes multiple clinical features.
- Age
- Anaemia
- Creatinine Phosphokinase (CPK)
- Diabetes
- Ejection Fraction
- High Blood Pressure
- Platelets
- Serum Creatinine
- Serum Sodium
- Sex
- Smoking
- Time (follow-up period)
- DEATH_EVENT → Indicates whether the patient died during the follow-up period (1 = Yes, 0 = No)
To build a machine learning model that predicts whether a patient is likely to experience a death event based on clinical features.
- Python 🐍
- Pandas
- NumPy
- Matplotlib
- Seaborn
- Scikit-learn
- Jupyter Notebook
- Load dataset using Pandas
- Explore structure and data types
- Check for missing values
- Feature selection
- Data cleaning (if required)
- Normalization / scaling (if applied)
- Distribution plots
- Correlation analysis
- Insights from clinical features
- Train machine learning models such as:
- Logistic Regression
- Decision Tree
- (Optional) Other models
- Accuracy score
- Confusion matrix
- Model performance comparison
The model is able to predict heart failure outcomes with good accuracy. This demonstrates the potential of machine learning in healthcare for early risk detection and decision support.