The National Institute for Occupational Safety and Health (NIOSH) announced a competition for programmers to develop artificial intelligence (AI) capable of analyzing safety reports and assigning occupational safety and health classification codes. Submissions are due by November 21.
When a worker is injured or becomes ill on the job, a person records free-form narrative text explaining how the injury or illness occurred. Currently, human evaluators read the reports and assign codes classifying injuries and illnesses. Reports can contain large amounts of information, and assigning codes is a costly and time-consuming task subject to human error.
NIOSH is asking programmers in the competition to develop algorithms using AI to automatically read injury records and classify them according to the Occupational Injury and Illness Classification System (OIICS).
NIOSH is working with vendor Topcoder, under an agreement with the National Aeronautics and Space Administration’s (NASA) Tournament Lab, to host the online competition.
“We’re thrilled to be hosting this competition along with our partners,” lead project officer Carlos Siordia, PhD, said in a NIOSH statement.
“Not only do these partnerships help support our extramural crowdsourcing AI competition, but they can also support others at CDC who want to crowdsource software programming to come up with the most innovative and efficient solutions to improve public health,” Siordia said.
Assigning OIICS Codes
NIOSH is interested in the potential for using AI to automatically read injury records and assign the appropriate OIICS codes.
The OIICS was developed by the Bureau of Labor Statistics (BLS) to provide a standardized coding system for characterizing work-related injuries and illnesses. OIICS codes reflect classifications for the nature of an injury or illness, body part affected, primary and secondary causes, and events or exposures involved.
The BLS uses the OIICS to characterize occupational fatalities in their annual Census of Fatal Occupational Injuries (CFOI) and nonfatal injuries and illnesses in the Survey of Occupational Injuries and Illnesses (SOII). The BLS last updated its coding classification manual in 2012.
The institute is providing programmers with a spreadsheet containing 229,820 injury and illness records. The spreadsheet contains injury and illness data in a comma-separated value (CSV) format.
NIOSH researchers randomly sampled 153,956 of the records with OIICS codes assigned to the event column of each record in a training set. For testing purposes, the remaining 75,864 records have not been assigned codes. Submitted algorithms should be able to learn from the training set how to assign codes to the test records.
Programmers may only use the Python and R programming languages. Prizes will be awarded for the top five entries. First prize is $10,000, second prize is $7,000, third prize is $5,000, fourth prize is $2,000, and fifth prize is $1,000.
NIOSH, a part of the Centers for Disease Control and Prevention (CDC), was created along with the Occupational Safety and Health Administration (OSHA) by the federal Occupational Safety and Health Act of 1970. NIOSH conducts research to develop new knowledge in the occupational safety and health field and to promote transfer of that knowledge into employer practices.