Skip to main content Start main content

A research team led by Dr Andy Cheng from RS and Department of Computing at PolyU has established the “Smart Work Injury Management System” (SWIM)

A team led by Dr Andy Cheng has established the “Smart Work Injury Management System” (SWIM)

A research team of Department of Rehabilitation Sciences and Department of Computing at PolyU has established the “Smart Work Injury Management System” (SWIM) to accurately predict with AI and big data the length of sick leaves and return-to-work trajectory in the cases of injuries at work.

Dr Andy Cheng, Associate Professor of RS, said they fed the system with around 90,000 cases collected from 68 insurance companies for building up analytic models through machine-learning. 

With the basic information such as age, job nature, salary, injured area and cause of injury, SWIM is able to predict the severity of a case, what treatment should be taken, the possibility of resumption of work, and the amount of insurance claim costs etc. 

SWIM’s prediction reaches 70% and 60% in accuracy to assess disability level and number of days of sick leave respectively, both outperforming estimations conducted by humans.

Dr Cheng said they hope SWIM’s prediction can be used as a reference for stakeholders like employers, injured workers and insurance companies to acquire a more comprehensive information so as to develop better rehabilitation plans and reduce the chances of work-related injuries.

Six insurance companies have already expressed their interest in SWIM. Dr Cheng said they expect to launch a 6-month trial run for these companies earliest by March, and put SWIM to use by the insurance and healthcare industries next year.


Online coverage:

HK01 -

Oriental Daily News -

Ming Pao Daily News -

Hong Kong Economic Times -

Sing Tao Daily -

am730 -

Sky Post -

Ta Kung Pao -

Wen Wei Po -

Economic Digest -

U Lifestyle -

Bastille Post -

6do News -


Your browser is not the latest version. If you continue to browse our website, Some pages may not function properly.

You are recommended to upgrade to a newer version or switch to a different browser. A list of the web browsers that we support can be found here