COVID-19 Patient Health Prediction Using Machine Learning

Overview This project leverages machine learning techniques to predict the health status of COVID-19 patients. The goal is to optimize resource allocation in healthcare settings by accurately forecasting patient outcomes based on their clinical and demographic data.

Key Features:

• Preprocessed data to handle missing and redundant entries.
• Implemented supervised and deep learning methods for classification.
• Conducted hyperparameter tuning to enhance model performance.
• Comprehensive analysis of feature importance and model results.

Data Overview

The dataset includes features such as:

• Demographics: Age, sex, location.
• Medical history: Chronic diseases, previous conditions.
• COVID-19 specific data: Symptom onset date, hospital admission date, lab results.
• Outcome Labels: Recovered, hospitalized, deceased, etc.

Data preprocessing involved:

• Removing or imputing missing data.
• Standardizing features for model compatibility. -Encoding categorical variables.

Methods

Supervised Learning Models

• Logistic Regression
• Decision Tree
• Random Forest
• Support Vector Machine
• K-Nearest Neighbors
• Naïve Bayes
• Ensemble methods (Adaboost, Bagging, Voting)

Deep Learning Models

• Feedforward Neural Network

Results

Performance Metrics:

Models were evaluated using:

• Accuracy
• Precision
• Recall
• F1-score

Highlights:

Adaboost with Decision Tree base classifier achieved the best performance. Extreme Gradient Boosting (XGBoost) showed robust results across all metrics. Feedforward Neural Networks were effective for complex data patterns.