Client’s Challenge
Patients with dementia are more expressive than healthy people when displaying pain. In the absence of self-report, facial expressions become an essential part of communicating the existence of pain, particularly for people with dementia. Facial expressions provide instant and brief signals to alert the observer. Facial descriptors are also valid indicators in pain observation scales for non-verbal patients with dementia. However, the reliability of observers in identifying these descriptors is often low, as the descriptors are generic, vague and cannot be recorded consistently.
Our Solution
To address the challenge these observers had, we've developed an innovative digital tool. Using artificial intelligence and supervised learning techniques, we've trained a model specifically to identify key facial descriptors indicative of pain. This model has been integrated into a mobile application, serving as an assistant for caregivers and medical professionals. By analyzing facial expressions, the app provides reliable pain assessments based on the facial descriptors and an established medical pain scales, enhancing care for (non-verbal) patients with dementia.
