JMIR Publications

ABSTRACT

Background:

Facial paralysis patients require long term follow up to monitor facial nerve function. The current standard of care for evaluating facial symmetry and movement uses validated clinician-scoring tools such as the House-Brackmann facial paralysis score or the electronic, clinician-graded facial function scale (eFACE). Quantitative tools have been developed for research use, such as the Emotrics application, a computer vision tool for facial paralysis assessment which analyzes facial features like oral commissure height and position from selected photographs. However, such tools are difficult to use in real time in clinic and cannot analyze videos of facial movement. To meet this need, we developed FaceADE, a novel iOS application leveraging TrueDepth technology to quantify facial movement.

Objective:

Benchmark FaceADE against ground truth physical measurements and current research tools to establish application accuracy.

Methods:

Facial measurements of lower lip commissure from 20 patients with facial paralysis and 5 healthy non-facial paralysis volunteers were taken. Measurements of movement and symmetry gathered from FaceADE were benchmarked against physical measurements using an open-source image analysis software (ImageJ).

Results:

Our pilot study shows FaceADE to be accurate in measuring lip commissure movement compared to ImageJ in both healthy volunteer and facial paralysis cohorts. Intraclass correlation of the healthy volunteer cohort was 0.975 (95% CI: 0.958-0.985, P<.001) and in the facial paralysis cohort was 0.914 (95% CI: 0.865-0.945, P<.001).

Conclusions:

Our proposed TrueDepth-based analysis enables clinicians to capture clinically accurate static and dynamic facial measurements in a live video-recorded format. In harnessing this technology, clinicians and patients may be able to quantify the degree of facial paralysis and recovery while monitoring changes in an accessible way.