Health Insurance Review and Assessment Service (HIRA) 's job-handling process will likely become faster than now.

HIRA said Thursday that it would apply the Artificial Intelligence Optical Character Recognition (AI OCR) system from next month to enhance its administrative efficiency.

AI OCR is a combination of machine learning and computer vision algorithms. These algorithms analyze document layout during pre-processing to pinpoint which information should be recorded.

So far, the agency has spent 46,000 hours a year for hand-written input of paper documents, image files, and other data collected from nursing institutions. However, introducing AI OCR will automate its job handling process, minimize errors, and accurately process the reception work, upgrading its reviewing administration.

HIRA strived hard to introduce the AI OCR system from January of last year to this past March.

Starting with the system design through benchmarking early-adopting agencies last January, it has collected opinions from related departments and operated a consultative body to draw up job process improvement plans.

HIRA broke down the manual work, automatically extracted data from four types of standard forms, including objection form, with high improvement effect, and connected them to the screening system, ensuring the accuracy of the work. It has also turned written screening references into data and implemented a function to classify types according to document characteristics automatically.

“The introduction of the AI OCR system has reduced the job handling time and increased the accuracy of the administrative process, laying a foundation for more efficient work,” said Choi Dong-jin, head of the information operation office. “Through digital innovation of health and medical administration like this, we will continue to make the public feel improved administrative service.”