Fractal tics may pave way to faster diagnosis

People with Tourette syndrome make brief and involuntary movements and sounds — repeatedly blinking, for example, or grunting. But though such 'tics' characterise this typically lifelong neurological disorder, which usually develops during childhood, there are also other kinds of tic disorders. To receive a diagnosis of Tourette syndrome, a child must have had movement and sound tics for at least a year. This means there's generally a long wait between the onset of tics and Tourette's being ruled in or out. 

But that may change: a new paper in the Journal of the Royal Society Interface describes a method that could potentially cut that year-long diagnosis time down to just a day. Payton Beeler at Washington University and colleagues report that the timing of tics of someone with Tourette's show a fractal pattern — an infinite pattern that is self-similar across different scales. Not only that, but they report that the degree to which tics follow fractal patterns can predict how severe their condition will become.   

This new work builds on the results of a 1998 paper, which recognised that Tourette tics tend to occur in bouts of several tics within a few seconds, separated by longer tic-free periods. That paper also noted that, over longer timescales, bouts that last several seconds recur in group episodes over the course of hours. The researchers behind this work felt that this was "suggestive, but not conclusive" evidence for a fractal process in the timing of tics.

Kevin J. Black, a neuroscientist specializing in tic disorders at Washington University, and a co-author of the new paper, was aware of this earlier research. As part of a major study on tics, he had also amassed nearly 1,000 minutes of video footage of 78 children with Tourette syndrome. These videos documented the timing of tics in two evaluation sessions, held a year apart. Black wanted to know whether there might be evidence of fractality in this large data-set. So, Beeler adapted techniques typically used to study chaos and fractality in engineering scenarios to explore whether, and to what extent, the timing of tics in these clips followed a fractal pattern. For each recording, she created a measure of how far the timings strayed from predictable, fractal intervals towards near-random chaotic timings, called a 'fractal dimension value'.

Beeler's analysis revealed that the timing of the children's tics was indeed fractal in nature. It also showed that the fractal dimension value for each child was the same in the second evaluation as it was in the first. What's more, a child's fractal dimension value varied according to the degree of their Tourette's severity, as assessed in the clinic. (It also varied in direct response to some tic suppression techniques that children were taught and asked to try in clinic.)

These findings suggest that a fractal dimension value might serve as an objective measure of tic severity. The researchers also hope that it might be possible to use this approach to diagnose Tourette syndrome immediately, during a child's initial evaluation, rather than after a 12-month wait. 

More work is now needed, though, to explore whether fractal dimension values can reliably differentiate between Tourette's and other types of tic disorders, which require different forms of treatment. The researchers speculate that this will indeed be the case, but they didn't analyse tic timing data from patients with other tic disorders in this study.   

If this approach could be used to diagnose Tourette syndrome immediately, this would clearly help patients, who could start receiving treatment immediately. It could also alleviate worry for the many children (and their parents) who develop some tics, and who are sent for an initial Tourette's evaluation, but whose tics subsequently disappear. Of course, it would also be interesting to know why the tics associated with Tourette's have a fractal pattern — and what that might reveal about its neurological underpinnings. 

Read the paper in full: https://doi.org/10.1098/rsif.2021.0742