Perk: A Shiny Web Application to Predict and Visualize Concentrations of Pharmaceuticals in the Aqueous Environment

Abstract

Predicting active pharmaceuticals ingredients (APIs) concentration in the environment using modelling approaches is an important aspect in the assessment of their environmental risk, especially for the APIs with no or limited analytical detection methods. However, handling, validating, and incorporating diverse datasets including APIs prescription/consumption data, metabolism, flow data, removal efficiency during wastewater treatment, dilution factor for the modelling is often laborious and time-consuming. The aim of this manuscript is to evaluate R/Shiny based tool, PERK, to facilitate automated modelling and reporting predicted concentration (PC) of a comprehensive set of APIs in different environmental matrices. PERK helped to calculate PC in wastewater influent, effluent, and river and compare with measured concentrations (MC) for five catchments located in England. Prediction accuracy (PA), the ratio between PC and MC, can be also generated with the tool. PERK provides consistent interactive user-interface, enabling user to visualise the results with limited programming knowledge.

Keywords: Water fingerprinting, Chemical mining, Decision support system, environmental monitoring, Risk assessment tool, Model-based evaluation