Identifying and analyzing peaks in high-resolution exposure profiles

The exposureProfiles R-package was developed for the analysis of high-resolution data from low-cost sensors used in exposure assessment studies.

A main advantage of using sensors for exposure assessment is that data is collected every few seconds or minutes, giving detailed insights in how exposure varies over time. Especially exposure peaks (moments when exposure is abnormally high) are of high interest since these can contribute to a large portion of the total exposure.

The amount and shape of the peaks in a measurement can give information about the exposure situation, especially when compared with similar measurements. For example, if two workers perform the same task, but one worker has much higher peaks than the other, that would give additional information for deploying control measures.

Comparing exposure profiles can easily be done manually when it is only a few measurements. However, when many measurements are collected, this quickly becomes impractical. For a more standardized and automated comparison, the exposureProfiles package was developed. In this package, profiles are analyzed by three steps. First, the overall amount of variation in the dataset is calculated using an accumulation curve. Second, exposure peaks are identified using one of two available methods. Last, the identified exposure peaks are analyzed to give summary statistics, which can be used for the comparisons.