Table 4 Performance of the Cox model and the random forest classification model. Cox model adjusted for age, sex, modified CCI, hypertension, diabetes, hyperlipidemia, temperature, relative humidity, PM2.5, NO2, and SO2 in the corresponding year. Random forest classification: age, sex, modified CCI, hypertension, diabetes, hyperlipidemia, temperature, relative humidity, PM2.5, CO, SO2, NO, NO2, NOx, and O3.

From: Long-term exposure to particulate matter was associated with increased dementia risk using both traditional approaches and novel machine learning methods

 

Cox model

Random forest

AUC

AUC

Accuracy

Year 1

0.79

0.76

0.70

Years 1–3

0.79

0.76

0.70

Years 1–5

0.79

0.76

0.70

Years 1–10

0.78

0.75

0.69

Back to article page