Abstract:
The Sustainable Development Goal (SGD) number 3 is focused on Good health and hygiene. Necessary measures need to be put in place to contribute toward the fulfilment of this goal. According to the World Health Organization (WHO), there have been rising cases of chronic respiratory disease especially in developing countries resulting to over four million death annually across the globe. In addition to the, COVID-19 pandemic some other highly contagious respiratory diseases include tuberculosis, chronic obstructive pulmonary disease (COPD), lung cancer, influenza among others. Most of the time it may be difficult to tell if around an infected person leading to avoidable transmissions. Therefore, this research project was proposed as a measure to control and minimize the spread of such diseases. This study aimed at developing an AI (Artificial Intelligence) powered system that can be able to monitor the risk of spread of respiratory diseases through a wearable device to contain such disease and reduce their effects. In this study, we are proposing a contactless system that is able to detect an environment with a high risk of respiratory disease contamination. The device applies Internet of Things (IoT) sensing technologies for data collection from the user's immediate surrounding environment, mainly proximity to others, cough sound and body temperature. The collected data is then processed on the embedded device and Machine Learning (ML) algorithms applied for analysis of data to predict a possible exposure to respiratory diseases. The user will then get alerts to take appropriate actions to avoid possible infections. The collected data is also sent to the things speak platform from time to time via cellular network for further analytics and future research. This is an improvement to existing solutions in which data analytics and machine learning are applied on the cloud. This was necessitated due to connectivity constraints in Africa, energy constraints, and the need for real-time response. We move AI capabilities from cloud to data source (edge devices) using TinyML.
