Machine learning for modeling mortality with respect to smog and ambient air temperature.

In January 2017 an excessive number of deaths (with respect to the corresponding period in 2016) was reported for Poland. Number of deaths observed in Bielanski Hospital in Warsaw in January 2017 shows a consistent increase when compared to the country-wide data. Analysis indicates that the excess visible in hospital data was due to respiratory system related deaths. We were able to statistically quantify this excess and build a numerical predictive model linking the number of deaths to ambient air temperature and air pollution in Warsaw. We were also able to estimate the number of deaths attributed to air pollution. [ABSTRACT FROM AUTHOR]