Continuous Physiological Signal Monitoring Using Wearables for the Early Detection of Infectious Diseases: A Review: An AI Perspective

In the chapter, biometrics and wearable technology are discussed as possibilities for detecting and monitoring infections due to pathogenic organisms such as viruses, bacteria, and fungi. In recent years, after the COVID-19 pandemic, wearable sensors have gained attention for their use in infectious disease detection. Comparatively, to conventional lab-based testing, wearable sensors offer high reliability, computational efficiency, and continuous monitoring of physiological activity. The chapter presents numerous studies using machine learning algorithms to analyze physiological data from wearable devices to find early disease symptoms, estimate epidemic patterns, and forecast illness outcomes. The chapter describes three phases of continuous physiological signal monitoring with wearables. First, the preprocessing methods were analyzed for data cleaning, transforming the data into different dimensions or formats, and preparing the data for analysis. The next phase of the process involves extracting features and selecting features to enhance data functionality and accessibility. During the third phase, machine learning, deep learning, and ensemble methods for disease classification were summarized. Continuous physiological signal monitoring by that wearable device allows early detection of infectious diseases such as COVID-19, malaria, tuberculosis, etc. Patients may benefit from early diagnosis and treatment, leading to improved outcomes. The chapter concludes that wearable devices and machine learning-based disease diagnosis will play a crucial role in the future in fighting such diseases.