- A logistic Regression model using sklearn
- Predicts whether the person's heart has failed or not using his/her clinical reports during the follow up period
| Feature | Explanation |
|---|---|
| Age | Age of the patient |
| Anaemia | Decrease of red blood cells or hemoglobin |
| High blood pressure | If a patient has hypertension |
| Creatinine phosphokinase (CPK) | Level of the CPK enzyme in the blood |
| Diabetes | If the patient has diabetes |
| Ejection fraction | Percentage of blood leaving |
| Sex | Woman or man |
| Platelets | Platelets in the blood |
| Serum creatinine | Level of creatinine in the blood |
| Serum sodium | Level of sodium in the blood |
| Smoking | If the patient smokes |
| Time | Follow-up period |
| DEATH_EVENT | If the patient died during the follow-up period |
The dataset can be found on kaggle.
- Binary Data pie plots:
-About 31% patients' heart failed during the follow up period.
-The dataset doesn't seem to be balanced between male and female patients.
- Distribution plots:
- Contributions of the features:
Sex, smoking, platelets, high blood pressure, diabetes, creatinine phosphokinase and anaemia doesnt contribute much as compared to other Features to the model.
- Accuracy on train and test data:
- Confusion Matrix:
