An Integrated Method to Monitor Indoor Air Quality Using IoT for Enhanced Health of COPD Patients

Respiratory morbidity and mortality are linked to outdoor air quality. People with COPD and other populations that are susceptible to outdoor air, such as indoor air quality and breathing health, are less well-known. Monitoring the indoor environment’s quality is essential as more people spend their time inside. Hospitalisations among people have increased recently as a result of respiratory conditions including COPD and asthma, which are both frequent illnesses. Utilising a sensor-based IoT system to monitor indoor pollution can be crucial for the management of certain chronic illnesses. Many academics suggested conducting a trial using an unobtrusive Internet of Things-based monitoring device to passively monitor the indoor environment in the homes of persons with asthma. The system tracks indoor particulate matter and carbon dioxide levels as well as homeowner interior activities like cleaning, cooking, smoking, ventilation time, and other factors. In this study, a system for alerting COPD patients whenever the air quality deteriorates and drops below a specific level has been developed. This method enables the patients to leave the area to avoid certain serious problems. Air Quality Index (AQI) is calculated from air quality variables like particulate matter-2.5 (PM2.5), particulate matter-10 (PM10), CO, CO2 AutoML tools. In the proposed work four distinct machine learning methods such as support vector machine (SVM), random forest regression (RFR), CatBoost regression (CR), Light Gradient Boosting Machine (Light GBM) have been utilized to determine the AQI. Light Gradient Boosting Machine is selected as an suitable model based on model metrics: accuracy, Recall, Precision and F1 score. An accuracy score of 97% with good precision, recall, and F1.Proposed work also supports that fine particulate matter (PM2.5) is vital in predicting AQI.