Objectives Functional neurological disorders are common, but there is a lack of objective tests for these conditions. Although accelerometry can distinguish functional from other tremor types, it is not routinely used at the bedside. Computer vision describes the processing of camera images by computer. It requires only ubiquitous hardware (e.g. smartphone, laptop) and standard clinical assessment, i.e. simple observation. We investigated computer vision to detect tremor distraction/entrainment in functional tremor.
Design Early results comparing computer analysis of video from a functional tremor and an essential tremor.
Methods 30 s (60 fps) video of extended forearm was recorded using a smartphone, for a functional tremor and an essential tremor patient. From 15 s, each participant tapped in time with a 3 Hz metronome using the contralateral hand (outside the video frame). Computing algorithms amplified the magnitude of video pixel movement and then measured the direction and size of pixel movement over time.
Results After the metronome onset, there was a marked change in video pixel movement for the functional tremor patient, with the frequency concentrating at 3 Hz, and this was statistically significant by linear discriminant analysis. There was no significant change in pixel movement after the metronome for the essential tremor patient (frequency remained 8–12 Hz).
Conclusions Smartphone video pixel movement can detect functional tremor entrainment, suggesting a possible new objective, bedside test.
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.