Background and aims
Assessment of pain in patients with advanced dementia is challenging due to impaired cognition and communication abilities. Our aim was to address this by developing a pain assessment system (PainChek®) that overcomes current challenges through the use of artificial intelligence (AI) and smart automation. 

Methods
We developed a system which includes a point-of-care application that uses AI to detect facial action units indicative of pain.  This data when combined with other non-facial indicators of pain collected using digital checklists, enables real-time computation of a cumulative pain score.  Our concept was initially supported through Alzheimer’s Australia grant in 2012. Following stakeholder consultations the first prototype was released in November 2013. This was further refined and validated against the Abbey Pain Scale in people with moderate to severe dementia. 
 
Results
Our research findings demonstrated the tool has excellent psychometric and clinimetric properties; the results of which were used to support its clearance as a Class 1 medical device in Australia (TGA) and Europe (CE mark) in July 2017. Since that time, PainChek® has been successfully implemented in over 70 Australian residential aged care facilities (RACFs) in Australia, with more than 450 users completing over 36,000 pain assessments. On April 29th, 2019 the Australian government announced $5m in funding to facilitate the implementation of PainChek® in RACFs across Australia and their 100,000 residents living with dementia. 

Conclusions
Our research has resulted in the successful development and validation of the world’s first point-of-care app utilizing AI to assist in the assessment of pain in patients with dementia. Technology, now owned by PainChek Ltd, has been successfully implemented into clinical practice around Australia.

02 - THE PAINCHEK (R) STORY: FROM CONCEPT TO AUSTRALIA-WIDE IMPLEMENTATION