Washington — Raymond van Venetië, a doctoral student in mathematics at the University of Amsterdam, is the winner of NIOSH’s first external crowdsourcing competition.
NIOSH challenged programmers to use artificial intelligence to improve ways “to automatically read injury records and classify them in occupational safety and health surveillance systems.” According to a Feb. 26 press release, van Venetië’s competition entry helped enhance injury classification accuracy to 90% – an increase over NIOSH’s baseline of 82%.
“Currently, whenever an employee is injured at work, humans write explanations of how the injury occurred, read all narratives and assign codes to classify injuries,” the release states. “This takes time, has costs associated with it and risks the potential for human error.”
NIOSH says it will work with van Venetië on a web tool for safety professionals who “have an interest in the classification of injury narratives.”
The competition featured 961 entries from 388 people representing 26 countries.
