Predicting Heart Disease with Logistic Regression: A Comprehensive Guide Using the UCI Dataset

Hello, fellow data science enthusiasts! In this article, we‘ll dive deep into the fascinating world of logistic regression and its application in predicting heart disease using the renowned UCI heart disease dataset. Whether you‘re a beginner looking to grasp the fundamentals or an experienced practitioner seeking to expand your knowledge, this guide will provide you with valuable insights and practical techniques to master logistic regression.

Understanding Logistic Regression

Before we delve into the specifics of the UCI dataset, let‘s take a moment to understand what logistic regression is and why it‘s a powerful tool in the data scientist‘s arsenal. Logistic regression is a statistical method used for predicting binary outcomes, such as the presence or absence of heart disease. Unlike linear regression, which is used for continuous outcomes, logistic regression is specifically designed to model the probability of an event occurring.

At its core, logistic regression estimates the relationship between a set of independent variables (also known as features or predictors) and a binary dependent variable (the outcome or target). It does this by fitting a logistic function, or sigmoid curve, to the data. The logistic function maps any real-valued number to a value between 0 and 1, representing the probability of the event occurring.

Exploring the UCI Heart Disease Dataset

Now that we have a basic understanding of logistic regression, let‘s turn our attention to the UCI heart disease dataset. This dataset is a widely used benchmark in the machine learning community and contains clinical information about patients, along with their heart disease status.

The dataset consists of 13 features, including:

1. Age
2. Sex
3. Chest pain type
4. Resting blood pressure
5. Serum cholesterol
6. Fasting blood sugar
7. Resting electrocardiographic results
8. Maximum heart rate achieved
9. Exercise-induced angina
10. Oldpeak (ST depression induced by exercise relative to rest)
11. Slope of the peak exercise ST segment
12. Number of major vessels colored by fluoroscopy
13. Thalassemia

The target variable indicates the presence of heart disease, with 0 representing no heart disease and 1 indicating the presence of heart disease.

Implementing Logistic Regression in Python

Now that we have a good understanding of the dataset, let‘s dive into the implementation of logistic regression using Python. We‘ll be using the popular scikit-learn library, along with pandas and NumPy for data manipulation and analysis.

First, let‘s load the dataset and perform some basic exploratory data analysis:

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns

data = pd.read_csv(‘heart_disease_uci.csv‘)
print(data.head())
print(data.info())
print(data.describe())

Next, let‘s preprocess the data by handling missing values (if any), encoding categorical variables, and scaling numerical features:

from sklearn.preprocessing import StandardScaler, OneHotEncoder
from sklearn.compose import ColumnTransformer
from sklearn.pipeline import Pipeline

numeric_features = [‘age‘, ‘trestbps‘, ‘chol‘, ‘thalach‘, ‘oldpeak‘]
categorical_features = [‘sex‘, ‘cp‘, ‘fbs‘, ‘restecg‘, ‘exang‘, ‘slope‘, ‘ca‘, ‘thal‘]

numeric_transformer = StandardScaler()
categorical_transformer = OneHotEncoder(handle_unknown=‘ignore‘)

preprocessor = ColumnTransformer(
    transformers=[
        (‘num‘, numeric_transformer, numeric_features),
        (‘cat‘, categorical_transformer, categorical_features)
    ])

clf = Pipeline(steps=[(‘preprocessor‘, preprocessor),
                      (‘classifier‘, LogisticRegression())])

We‘ll split the data into training and testing sets and train our logistic regression model:

from sklearn.model_selection import train_test_split

X = data.drop(‘target‘, axis=1)
y = data[‘target‘]

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

clf.fit(X_train, y_train)

Finally, let‘s evaluate the model‘s performance on the testing set:

from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score

y_pred = clf.predict(X_test)

print(f"Accuracy: {accuracy_score(y_test, y_pred):.2f}")
print(f"Precision: {precision_score(y_test, y_pred):.2f}")
print(f"Recall: {recall_score(y_test, y_pred):.2f}")
print(f"F1-score: {f1_score(y_test, y_pred):.2f}")

Interpreting the Model‘s Coefficients

One of the great advantages of logistic regression is its interpretability. We can examine the model‘s coefficients to understand the impact of each feature on the prediction of heart disease. A positive coefficient indicates that the feature increases the probability of heart disease, while a negative coefficient suggests a decrease in probability.

Let‘s print the model‘s coefficients and their corresponding features:

coefs = pd.concat([pd.DataFrame(X.columns),pd.DataFrame(np.transpose(clf.named_steps[‘classifier‘].coef_))], axis = 1)
print(coefs)

Comparing Logistic Regression with Other Algorithms

While logistic regression is a powerful and interpretable algorithm, it‘s always a good idea to compare its performance with other classification algorithms. Let‘s train a decision tree, random forest, and support vector machine on the same dataset and evaluate their performance:

from sklearn.tree import DecisionTreeClassifier
from sklearn.ensemble import RandomForestClassifier
from sklearn.svm import SVC

classifiers = {
    ‘Logistic Regression‘: LogisticRegression(),
    ‘Decision Tree‘: DecisionTreeClassifier(),
    ‘Random Forest‘: RandomForestClassifier(),
    ‘Support Vector Machine‘: SVC()
}

for clf_name, clf in classifiers.items():
    pipeline = Pipeline(steps=[(‘preprocessor‘, preprocessor),
                               (‘classifier‘, clf)])
    pipeline.fit(X_train, y_train)
    y_pred = pipeline.predict(X_test)
    print(f"{clf_name} - Accuracy: {accuracy_score(y_test, y_pred):.2f}")

Limitations and Future Directions

While logistic regression is a valuable tool for predicting heart disease, it‘s important to be aware of its limitations. One potential issue is multicollinearity, which occurs when the independent variables are highly correlated with each other. This can lead to unstable and unreliable coefficients. Additionally, logistic regression assumes a linear relationship between the independent variables and the log odds of the outcome, which may not always hold true.

Future directions for research could include exploring techniques for handling imbalanced datasets, where the number of instances in each class is significantly different. Oversampling the minority class or undersampling the majority class are common approaches to address this issue. Another area of interest is feature selection, which involves identifying the most informative features for predicting heart disease and potentially reducing the dimensionality of the dataset.

Conclusion

In this article, we‘ve explored the power of logistic regression for predicting heart disease using the UCI dataset. We‘ve covered the fundamentals of logistic regression, preprocessed the data, trained and evaluated our model, interpreted the coefficients, and compared its performance with other classification algorithms.

Logistic regression is a valuable tool in the healthcare domain, enabling medical professionals to make informed decisions and prioritize patient care based on the predicted likelihood of heart disease. By understanding the factors that contribute to heart disease, we can develop targeted interventions and preventive measures to improve patient outcomes.

Remember, while logistic regression is a powerful technique, it‘s just one of many tools in the data scientist‘s toolbox. Continuously exploring new algorithms, techniques, and approaches is essential to staying at the forefront of the field and making meaningful contributions to healthcare and beyond.

Thank you for joining me on this journey through logistic regression and the UCI heart disease dataset. I hope you found this guide informative and engaging. Until next time, happy data sciencing!