Background: 
Toe walking is a clinical motor sign present in 20% of individuals with autism spectrum disorder (ASD). Because this behavior is also found during standing or running, the term tip-toe behavior (TTB) seems to be more appropriate. A systematic review found a lacking of studies that quantify TB in individuals with ASD. The most used and reliable instrumental tool to quantify motor deficit during standing and walking is the gait analysis, but this approach requests the positioning of a large number of markers on the skin of the patient, a long time to prepare the patient for the execution of the test and is conducted in a non-ecological setting. For all these reasons gait analysis could be difficult to use with individuals with severe ASD because of their clinical condition. Thus, a simpler instrumental approach that uses wearable sensors usable in an ecological setting could be a useful resource to quantify TTB in individuals with severe ASD. 

Objectives: 
The aims of this pilot study are: 1) to verify the feasibility and acceptability of wearable sensors in individuals with severe ASD; 2) to quantify TB using wearable sensors during structured standing and walking tasks. 

Methods: 
Subjects with ASD diagnosed according to DSM-5 criteria and confirmed using the Autism Diagnostic Observation Schedule were admitted to the study. TB was quantified using "Sensoria® Smart Socks" (SSS), a commercially available wireless gait monitoring technology (Figure 1). SSS was found a valid measure instrument in a previous study. To assess the TTB quantity during standing and walking we used a static and dynamic test following the methodology proposed in a previous study. The static test consisted in playing while standing in front of a table for 3 minutes. The dynamic test consisted in transporting 1 object (puzzle, Lego®, ...) from the therapist to the playing table situated 2 meters away and back again 15 times. The test is conducted without shoes albeit with SSS. The same person was tested three times on three different days (9 acquisitions). The data acquired from the SSS were elaborated from a dedicated algorithm. The result of the dynamic test is presented as the mean percentage number of toe steps and the result of the static test is reported as the mean percentage of seconds in TB. 

Results: 
Three individuals with severe ASD and TTB were assessed. Their age was 10.9yrs, 12.8yrs and 13yrs (3/3 males) and their ADOS CSS was 9, 10 and 8, respectively. We found the SSS was feasible and acceptable in the three individuals with severe ASD and TTB in all three trials. We were also able to quantify TTB during both the static and the dynamic tests using the SSS tool in 9/9 of the acquisition (100%) (example in Figure 2). 

Conclusions: 
"Sensoria® Smart Socks" seems to be a feasible and acceptable wearable sensors approach for quantifying TTB in individuals with severe ASD and TTB. Further research is required to confirm these preliminary results.

Notes:

1 - Villa Santa Maria Foundation Autism Research Unit, Tavernerio (Como), Italy

2 - Politecnico di Milano, Italy