Predicting Heart Disease Risk with Machine Learning Models in Python¶
This project leverages machine learning techniques to develop a predictive system for diagnosing heart disease based on key medical features. By utilizing a dataset of heart disease patients, the system employs a Random Forest classifier to predict whether an individual is at risk of heart disease. The model is designed to assist healthcare professionals by providing accurate, data-driven predictions that can aid in early diagnosis and decision-making.
Key Features:¶
Data Preprocessing: The dataset undergoes preprocessing to handle missing values, encode categorical data, and scale numerical features to ensure high model accuracy.
Model Development: A Random Forest classifier is trained to predict heart disease risk, and model performance is evaluated using accuracy scores, classification reports, and confusion matrices.
Feature Importance Analysis: An analysis of the most significant features contributing to the model's predictions helps in understanding which medical parameters are most indicative of heart disease.
Prediction for New Data: The system can make predictions for new patients based on their medical records.
Visualization of Results: Prediction probabilities for "Heart Disease" and "No Heart Disease" are visualized with bar charts to provide a clear understanding of the model’s confidence.
Conclusion:¶
This Heart Disease Prediction System uses machine learning to provide an early warning for individuals at risk of heart disease, supporting timely intervention and better healthcare outcomes. With clear visualizations and an interpretable model, this system can be a valuable tool for both healthcare providers and researchers.
In [6]:
# Install Required Packages
!pip install pandas numpy matplotlib seaborn scikit-learn joblib
Requirement already satisfied: pandas in c:\users\omprakash\anaconda3\lib\site-packages (2.2.2) Requirement already satisfied: numpy in c:\users\omprakash\anaconda3\lib\site-packages (1.26.4) Requirement already satisfied: matplotlib in c:\users\omprakash\anaconda3\lib\site-packages (3.9.2) Requirement already satisfied: seaborn in c:\users\omprakash\anaconda3\lib\site-packages (0.13.2) Requirement already satisfied: scikit-learn in c:\users\omprakash\anaconda3\lib\site-packages (1.5.1) Requirement already satisfied: joblib in c:\users\omprakash\anaconda3\lib\site-packages (1.4.2) Requirement already satisfied: python-dateutil>=2.8.2 in c:\users\omprakash\anaconda3\lib\site-packages (from pandas) (2.9.0.post0) Requirement already satisfied: pytz>=2020.1 in c:\users\omprakash\anaconda3\lib\site-packages (from pandas) (2024.1) Requirement already satisfied: tzdata>=2022.7 in c:\users\omprakash\anaconda3\lib\site-packages (from pandas) (2023.3) Requirement already satisfied: contourpy>=1.0.1 in c:\users\omprakash\anaconda3\lib\site-packages (from matplotlib) (1.2.0) Requirement already satisfied: cycler>=0.10 in c:\users\omprakash\anaconda3\lib\site-packages (from matplotlib) (0.11.0) Requirement already satisfied: fonttools>=4.22.0 in c:\users\omprakash\anaconda3\lib\site-packages (from matplotlib) (4.51.0) Requirement already satisfied: kiwisolver>=1.3.1 in c:\users\omprakash\anaconda3\lib\site-packages (from matplotlib) (1.4.4) Requirement already satisfied: packaging>=20.0 in c:\users\omprakash\anaconda3\lib\site-packages (from matplotlib) (24.1) Requirement already satisfied: pillow>=8 in c:\users\omprakash\anaconda3\lib\site-packages (from matplotlib) (10.4.0) Requirement already satisfied: pyparsing>=2.3.1 in c:\users\omprakash\anaconda3\lib\site-packages (from matplotlib) (3.1.2) Requirement already satisfied: scipy>=1.6.0 in c:\users\omprakash\anaconda3\lib\site-packages (from scikit-learn) (1.13.1) Requirement already satisfied: threadpoolctl>=3.1.0 in c:\users\omprakash\anaconda3\lib\site-packages (from scikit-learn) (3.5.0) Requirement already satisfied: six>=1.5 in c:\users\omprakash\anaconda3\lib\site-packages (from python-dateutil>=2.8.2->pandas) (1.16.0)
In [8]:
import pandas as pd
import numpy as np
import seaborn as sns
import matplotlib.pyplot as plt
from sklearn.model_selection import train_test_split
from sklearn.ensemble import RandomForestClassifier
from sklearn.preprocessing import StandardScaler, LabelEncoder
from sklearn.metrics import accuracy_score, classification_report, confusion_matrix, ConfusionMatrixDisplay
import joblib
In [10]:
# Load the Heart Disease dataset
heart_data = pd.read_csv("heart.csv")
In [12]:
# Display the first few rows of the dataset
print("First few rows of Heart Disease dataset:")
print(heart_data.head())
First few rows of Heart Disease dataset: age sex cp trestbps chol fbs restecg thalach exang oldpeak slope \ 0 52 1 0 125 212 0 1 168 0 1.0 2 1 53 1 0 140 203 1 0 155 1 3.1 0 2 70 1 0 145 174 0 1 125 1 2.6 0 3 61 1 0 148 203 0 1 161 0 0.0 2 4 62 0 0 138 294 1 1 106 0 1.9 1 ca thal target 0 2 3 0 1 0 3 0 2 0 3 0 3 1 3 0 4 3 2 0
In [14]:
# Step 1: Data Preprocessing
# Handle missing values (if any)
heart_data.fillna(heart_data.mean(), inplace=True)
In [16]:
# Step 2: Encoding the target variable 'target' to binary (0: No heart disease, 1: Heart disease)
encoder = LabelEncoder()
heart_data['target'] = encoder.fit_transform(heart_data['target'])
In [18]:
# Step 3: Feature Scaling
# Separate features and target
X = heart_data.drop('target', axis=1) # Features
y = heart_data['target'] # Target
In [20]:
# Normalize the features
scaler = StandardScaler()
X_scaled = scaler.fit_transform(X)
In [22]:
# Step 4: Train-Test Split (80% train, 20% test)
X_train, X_test, y_train, y_test = train_test_split(X_scaled, y, test_size=0.2, random_state=42)
In [24]:
# Step 5: Train the Random Forest Classifier
rf_model = RandomForestClassifier(random_state=42)
rf_model.fit(X_train, y_train)
Out[24]:
RandomForestClassifier(random_state=42)In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.
RandomForestClassifier(random_state=42)
In [26]:
# Step 6: Make Predictions
y_pred = rf_model.predict(X_test)
In [28]:
# Step 7: Evaluate the Model
# Accuracy
print(f"Accuracy of Random Forest Model: {accuracy_score(y_test, y_pred):.2f}")
Accuracy of Random Forest Model: 0.99
In [30]:
# Classification report
print("\nClassification Report:")
print(classification_report(y_test, y_pred))
Classification Report:
precision recall f1-score support
0 0.97 1.00 0.99 102
1 1.00 0.97 0.99 103
accuracy 0.99 205
macro avg 0.99 0.99 0.99 205
weighted avg 0.99 0.99 0.99 205
In [32]:
# Confusion Matrix Visualization
cm = confusion_matrix(y_test, y_pred)
disp = ConfusionMatrixDisplay(confusion_matrix=cm, display_labels=encoder.classes_)
disp.plot(cmap="Blues")
plt.title("Confusion Matrix for Heart Disease Model")
plt.show()
In [34]:
# Step 8: Feature Importance Visualization
def plot_feature_importance(model, feature_names, title="Feature Importance"):
feature_importances = model.feature_importances_
sorted_idx = np.argsort(feature_importances)[::-1]
plt.figure(figsize=(10, 6))
sns.barplot(x=feature_importances[sorted_idx], y=feature_names[sorted_idx])
plt.title(title)
plt.xlabel('Importance')
plt.ylabel('Features')
plt.show()
In [36]:
# Plot Feature Importance
plot_feature_importance(rf_model, X.columns, "Feature Importance for Heart Disease Model")
# Step 9: Save the Model and Scaler
joblib.dump(rf_model, "heart_disease_model.pkl")
joblib.dump(scaler, "scaler.pkl")
print("\nModel and Scaler saved!")
Model and Scaler saved!
In [38]:
# Step 10: Load the saved model and scaler for prediction
loaded_model = joblib.load("heart_disease_model.pkl")
loaded_scaler = joblib.load("scaler.pkl")
In [40]:
def predict_new_data(model, scaler, new_data, feature_names):
# Convert the new data to a DataFrame to align with the feature names
new_data_df = pd.DataFrame([new_data], columns=feature_names)
# Scale the new data
scaled_input = scaler.transform(new_data_df)
# Make the prediction
prediction = model.predict(scaled_input)
# Interpret the prediction (0: No heart disease, 1: Heart disease)
return "No Heart Disease" if prediction[0] == 0 else "Heart Disease"
In [42]:
# Use the feature names (column names of the Heart Disease dataset)
feature_names = X.columns.tolist()
# Example input data for a new patient
new_patient_data = [63, 145, 233, 1, 150, 0.0, 2, 174, 0, 0, 0, 3, 0.0]
# Make prediction for the new patient
prediction = predict_new_data(loaded_model, loaded_scaler, new_patient_data, feature_names)
print(f"Prediction for New Patient: {prediction}")
Prediction for New Patient: Heart Disease
In [44]:
import matplotlib.pyplot as plt
# Example new patient data for two cases:
new_patient_with_heart_disease = [63, 145, 233, 1, 150, 0.0, 2, 174, 0, 0, 0, 3, 0.0]
new_patient_without_heart_disease = [60, 130, 230, 0, 120, 0.0, 1, 160, 0, 0, 0, 2, 0.0,]
In [46]:
# Get probabilities for both cases (heart disease vs. no heart disease)
prob_with_heart_disease = loaded_model.predict_proba([new_patient_with_heart_disease])[0]
prob_without_heart_disease = loaded_model.predict_proba([new_patient_without_heart_disease])[0]
In [48]:
# Create a bar plot to compare the two cases
labels = ['No Heart Disease', 'Heart Disease']
fig, axes = plt.subplots(1, 2, figsize=(12, 6))
# Plot for the patient with heart disease
axes[0].bar(labels, prob_with_heart_disease, color=['blue', 'red'])
axes[0].set_title('Prediction Probability: Patient with Heart Disease')
axes[0].set_ylim(0, 1)
axes[0].set_xlabel('Prediction')
axes[0].set_ylabel('Probability')
# Plot for the patient without heart disease
axes[1].bar(labels, prob_without_heart_disease, color=['blue', 'red'])
axes[1].set_title('Prediction Probability: Patient without Heart Disease')
axes[1].set_ylim(0, 1)
axes[1].set_xlabel('Prediction')
axes[1].set_ylabel('Probability')
# Display the plots
plt.tight_layout()
plt.show()
