The rapid and accurate detection of infectious diseases like COVID-19 is critical for timely intervention, especially in asymptomatic or early-stage cases. This project, titled "Wearable Design for Infectious Disease Detection Through Machine Learning," aims to leverage data from wearable health monitoring devices to develop an intelligent, automated system for detecting infectious conditions. The dataset used comprises both cross-sectional and time-series physiological signals such as body temperature, heart rate, oxygen saturation (SpO₂), respiratory rate, heart rate variability (HRV), and pulse pressure, with the target variable being a binary COVID-19 diagnosis label (0: Negative, 1: Positive).

The system integrates a variety of machine learning and deep learning models tailored for both static and sequential data. Traditional classifiers like Random Forest, Gradient Boosting, AdaBoost, and Extra Trees are employed for cross-sectional prediction, with a soft-voting ensemble strategy to enhance overall robustness and accuracy. For deep sequence modeling, Long Short-Term Memory (LSTM) networks and 1D Convolutional Neural Networks (CNNs) are utilized to capture temporal dependencies within the health signals. Additionally, Time Series Forest (TSF) is applied for modeling the temporal dynamics directly from the time-series data.

This hybrid approach allows for comprehensive pattern recognition from both static health snapshots and continuous streams of wearable data. The proposed system demonstrates high potential for real-time, non-invasive, and continuous monitoring of infectious diseases using low-cost wearable technology, promoting proactive healthcare and early-stage outbreak management.

Keywords: infectious Disease Detection, Wearable Health Monitoring, Time-Series Analysis, Cross-Sectional Modeling, Random Forest, Gradient Boosting, AdaBoost, Extra Trees Classifier, Voting Classifier, LSTM, CNN, Time Series Forest, Physiological Signal Analysis, COVID-19 Prediction, Machine Learning, Deep Learning, Biomedical Signal Processing, Health Informatics.