Wildfires emit large quantities of air pollutants into the atmosphere. As wildfires increase in frequency, intensity, duration, and coverage area, such emissions have become a significant health hazard for residential populations, particularly the vulnerable groups. This health hazard is exacerbated by two factors: first, wildfires are expected to increase in frequency as a result of climate change; and second, fine particulate matter, PM2.5, in wildfire smoke adversely impacts human health. Recent toxicological studies suggest that wildfire particulate matter may be more toxic than equal doses of ambient PM2.5. We will present how to forecast the human health burden of wildfire emissions via deep learning models. We are developing a novel statistical framework for forecasting future emissions from active wildfires by integrating physicochemical models of emissions and satellite observations with statistical forecasting models. This is allowing us to model human health impacts of poor air quality, and use this to forecast the burden of diseases associated with exposures to wildfire events, both short- and long-term, and help develop mitigation strategies.