=Paper=
{{Paper
|id=Vol-2524/paper25
|storemode=property
|title=Wearable devices to help children with autism overcome toe walking
|pdfUrl=https://ceur-ws.org/Vol-2524/paper25.pdf
|volume=Vol-2524
|authors=Andrea Mennitto,Massimiliano Caretti,Luigi Iovino
|dblpUrl=https://dblp.org/rec/conf/psychobit/MennittoCI19
}}
==Wearable devices to help children with autism overcome toe walking==
https://ceur-ws.org/Vol-2524/paper25.pdf
Wearable devices to help children with autism overcome
toe walking
Andrea Mennitto1, Massimiliano Caretti2 and Luigi Iovino1
1 Neapolisanit S.R.L. Rehabilitation Center, Ottaviano NA 80044, Italy
2 Italian National Research Council, Rome, Italy
a.mennitto@neapolisanit.net
Abstract. One of the first symptoms of Autism Spectrum Disorder is "Toe-Walk-
ing", a walking mode in which the person uses only the tip of the foot, without
bringing the heel to the ground. Common in toddlers, children with ASD continue
to use this pattern growing up. Toe Walking can lead to the development of sec-
ondary motor deformity, the shortening of a heel cord, whose treatment may re-
quire surgery (eg. tendon lengthening).
Aim of this study is to analyze behavioral interventions, which are non-invasive
and easily integrated with rehabilitation activities already in place in the centers
and homes of many young people with ASD. The research team will then present
a proposal for a technological evolution on this issue, in order to develop a system
for early diagnosis and treatment.
Keywords: autism, gait, toe walking.
1 Introduction
Autism is a neuropsychological condition that can be diagnosed in the first years of life
and accompanies the individual throughout his life [1]. Autism is often associated with
hypo or hypersensitivity to sensory stimuli, often with numerous and unpredictable shift
between the two [2].
This altered sensitivity could also be one of the causes of toe-walking, a stereotypic
behavior frequently adopted by individuals with Autism Spectrum Disorder [3]. Toe-
walking is a term used to describe “the condition in which children walk with a toe-toe
gait pattern in the absence of any known cause” [4]. There are no definite data on the
prevalence of idiopathic toe-walking, but some studies record this type of behavior in
20% of subjects with Autism Spectrum Disorder [5].
Although a direct cause-effect relationship has not been identified, toe-walking often
accompanies difficulty in developing language. One study in particular highlights the
association between toe-walking and language difficulties also in subjects without a
diagnosis of neurodevelopmental problems [6].
Often, toe-walking in autism can lead to a secondary motor deformity, who needs to
be medically treated, in particular with stretching exercises, casting, braces, Botox and
2
heel cord surgery [7]. Early intervention is the key to avoiding more invasive proce-
dures.
In recent years, advances in technology have led to the development of increasingly
complex applications for the habilitation and the rehabilitation of subjects with ASD.
Hardware and software came together [8] in order to guarantee therapeutic advance-
ments in fields such as communication [9], leading for a future of convergence where
the machine facilitates, not monopolizes, the opportunities for growth of the individual.
The purpose of this paper is to highlight the joint efforts between technology and be-
havioral sciences [10], trying to develop an intervention model for toe-walking in the
Autistic Spectrum Disorder.
2 Interventions for toe walking in autism
2.1 TAGteach
Teaching with acoustical guidance, or TAGteach [11], is a behavioral intervention
based on the use of clickers as a conditioned reinforcer. It uses a conditioned auditory
stimulus (a "click" made by a low tech device) as a positive reinforcement to correct
behaviors. The “click” is paired with an already established reinforcing item, in order
to increase positive behaviors.
Despite the empirical evidence on the effectiveness of the intervention, the available
studies are scarce and on a small number of people.
A study published in 2014 by Persicke, Jackson, and Adams [11], evaluated the ef-
fectiveness of the intervention on a 4-year-old child with a diagnosis of Autism Spec-
trum Disorder. The study was divided into the following phases: baseline (A), correc-
tion only (B), and correction + TAG phase (BC). At the end of each phase a reversal
(A) was performed, identical to the baseline phase, and the intervention ended with two
generalization probes to novel environments. Five sessions a day were performed for
2-5 days each week. The sessions were videotaped and subsequently analyzed to ex-
trapolate data on the walk. The correction phase involved the physical intervention of
the therapist, while the subsequent one aimed to increase the number of correct steps,
modifying the reinforcement ratio. Once the intervention was concluded, the child had
gone from an average of 24.6% steps performed correctly to a percentage higher than
73%.
2.2 GaitSpot
GaitSpot Auditory Shoe Squeakers [12] are a low-tech device to stick inside or outside
shoes. On this accessory is mounted a squeaker in correspondence with the heel. By
compressing this squeaker between the heel and the ground, it is possible to hear im-
mediate sound feedback, working both as a stimulus prompt and as an auditory conse-
quence.
A study published by Marcus et al. in 2010 investigated the effectiveness of the sim-
plified habit reversal training procedure and differential reinforcement of incompatible
3
behavior using GaitSpot with three boys, aged between 8 and 9 years, with a diagnosis
of autism. A delayed multiple baseline with changing criteria across participants design
was used. The intervention, videotaped, was carried out 5 days a week, inside a special
school attended by all three children.
Before the procedure, the squeakers were paired with small food reinforcements or
tokens. The procedure included multiple baseline sessions (lasting 10 minutes), in
which the children walked freely. In the training phase, the child's walk was accompa-
nied by vocal verbal prompt, with the reinforcement delivered every three steps per-
formed correctly. After ten minutes of walking without toe-walking, the researchers
started the next phase.
Phase C, lasting 10 minutes per treatment session, differs from the previous one due
to the absence of verbal vocal prompts and the VR5 reinforcement ratio. Phase D, on
the other hand, involves the use of GaitSpots for the entire school day, with the prompt
given only once at the beginning of the recording.
During phase E the recording took place when the children did not wear GaitSpots,
while in the following ones the less controlled environment and the fading of the aids
determined the generalization of the newly acquired behavior.
The results of the study were positive, with the mean percentage of target behaviour
in all phases “showing a substantial decrease in habitual toe-walking behaviour from
baseline”, although the limited sample and the difficulties in intervening outside the
school context led to a lack of generalization.
2.3 CARE.ME Walk
Although the methodology used, strongly based on the analysis of behavior, provides
interesting application cues, the low-tech approach of current studies is one of the rea-
sons why the aforementioned works are difficult to apply to heterogeneous and numer-
ous populations. The need to create a baseline for each user, the request for 1:1 treat-
ment and the dependence on devices that are too simple for real-time reporting require
extra effort from therapists, parents and above all young patients. At the same time,
high-tech, reliable but cheap wearable devices are available on the market today. With
these devices, it is possible to partially solve these problems and enhance the work of
the clinician.
In order to build a prototype, a Tactigon ONE [13] device was used. The device,
small and easy to place on patients' shoes, is equipped with 3 axis gyroscope and 3 axis
accelerometer and can communicate via Bluetooth to a device such as a smartphone or
to a small receiver in the environment, bridging physical and digital world together
[14].
The data, collected and processed in real time using a proprietary algorithm [15]
trough an Android smartphone, can be used to trace the movements of the foot, measure
the three angles of rotation (pitch, roll, and yaw) and identify regular walking from toe-
walking. The app that collects the data will present the patient with a summary of what
has been measured, using a gamified environment [16] capable of motivating [17] the
person to reach the next goal. The app interface is personalized with an avatar capable
4
of guiding the subject in the training up to the final stage, in which the functional be-
havior will be reinforced with digital tokens, redeemable with rewards in the real world
with the help of parents, teachers, and caregivers.
The feedback, managed locally by the device, can consist of customizable sounds or
a slight vibration, in order to provide information to the patient without becoming a
source of stress [18].
An ad-hoc cloud platform [15] takes care of storing data, performing advanced sta-
tistical processing and allowing an overview of the work performed both by clinicians
and researchers involved in the project.
3 Conclusions
Toe-walking is a known symptom of autism which, if not treated properly and at the
right time, often leads to serious effects for the person. Some behavioral protocols,
combined with low-tech tools such as the aforementioned TAGteach and GaitSpot,
have shown promising results, paving the way for new research.
Today it is possible to tap into the market of wearable devices in order to have tech-
nologically advanced sensors at reasonable prices, and modern smartphone calculation
skills allow data processing compatible with real-time feedback results.
Our prototype makes it possible to enhance the consolidated behaviorist approach,
automating expensive processes such as data analysis and allowing constant use, even
in unsupervised environments.
References
1. American Psychiatric Association: Diagnostic and statistical manual of mental disorders
(DSM-5®). American Psychiatric Pub (2013).
2. O’Neill, M., Jones, R.S.P.: Sensory-perceptual abnormalities in autism: A case for more
research? J. Autism Dev. Disord. 27, 283–294 (1997).
https://doi.org/10.1023/A:1025850431170.
3. Klintwall, L., Holm, A., Eriksson, M., Carlsson, L.H., Olsson, M.B., Hedvall, Å.,
Gillberg, C., Fernell, E.: Sensory abnormalities in autism. A brief report. Res. Dev.
Disabil. 32, 795–800 (2011). https://doi.org/10.1016/j.ridd.2010.10.021.
4. Disorders, D., Diseases, J., Kennedy, R.F.: Idiopathic toe-walking: a review. 846–848
(1999).
5. Ming, X., Brimacombe, M., Wagner, G.C.: Prevalence of motor impairment in autism
spectrum disorders. Brain Dev. 29, 565–570 (2007).
6. Heaney, M.S., Whitman, B., Accardo, P., Morrow, J., Tomazic, T.: Toe Walking and
Language Development. Clin. Pediatr. (Phila). 31, 158–160 (2007).
https://doi.org/10.1177/000992289203100306.
7. Barrow, W.J., Jaworski, M., Accardo, P.J.: Persistent toe walking in autism. J. Child
Neurol. 26, 619–621 (2011). https://doi.org/10.1177/0883073810385344.
5
8. Ponticorvo, M., Di Fuccio, R., Ferrara, F., Rega, A., Miglino, O.: Multisensory
Educational Materials: Five Senses to Learn. In: International Conference in
Methodologies and intelligent Systems for Techhnology Enhanced Learning. pp. 45–
52. Springer (2018).
9. Rega, A., Mennitto, A., Iovino, L.: Liar (Language Interface for Autistic’S
Rehabilitation): Technological Aids for Specialists Supporting the Acquisition of
Verbal Behavior in Persons With Autism. EDULEARN17 Proc. 1, 1755–1760 (2017).
https://doi.org/10.21125/edulearn.2017.1375.
10. Ponticorvo, M., Rega, A., Miglino, O.: Toward Tutoring Systems Inspired by Applied
Behavioral Analysis. In: International Conference on Intelligent Tutoring Systems. pp.
160–169. Springer (2018).
11. Persicke, A., Jackson, M., Adams, A.N.: Brief report: An evaluation of TAGteach
components to decrease toe-walking in a 4-year-old child with autism. J. Autism Dev.
Disord. 44, 965–968 (2014). https://doi.org/10.1007/s10803-013-1934-4.
12. Marcus, A., Sinnott, B., Bradley, S., Grey, I.: Treatment of idiopathic toe-walking in
children with autism using GaitSpot Auditory Speakers and simplified habit reversal.
Res. Autism Spectr. Disord. 4, 260–267 (2010).
https://doi.org/10.1016/j.rasd.2009.09.012.
13. Products - The Tactigon, https://www.thetactigon.com/products/.
14. Miglino, O., Di Ferdinando, A., Di Fuccio, R., Rega, A., Ricci, C., Schembri, M., Ricci,
C.: Bridging Digital and Physical Educational Games Using RFID/NFC Technologies.
J. E-Learinig Knowl. Soc. 3, 87–104 (2014).
15. Rega, A., Iovino, L., Somma, F., Mennitto, A., Granata, A.: Automating Data Collection
and Analysis in Psychological and Educational Treatments for Autism. ICERI2018
Proc. 1, 5502–5507 (2018). https://doi.org/10.21125/iceri.2018.2277.
16. Ponticorvo, M., Rega, A., Di Ferdinando, A., Marocco, D., Miglino, O.: Approaches to
embed bio-inspired computational algorithms in educational and serious games. CEUR
Workshop Proceedings, 2099, 8-14, (2018).
17. Rega, A., Mennitto, A., Vita, S., Iovino, L.: New Technologies And Autism: Can
Augmented Reality (AR) Increase The Motivation In Children With Autism? In:
INTED2018 Proceedings. pp. 4904–4910 (2018).
https://doi.org/10.21125/inted.2018.0959.
18. Rega, A., Mennitto, A.: Augmented Reality As An Educational And Rehabilitation
Support For Developmental Dyslexia. In: ICERI2017 Proceedings. pp. 6969–6972
(2017).