Rehabilitation of facial Praxia in Cerebral palsy using interactive Avatar – RePliCA

RePliCA project is dedicated to children suffering from cerebral palsy (CP). Cerebral palsy is the first cause of motor troubles in children. Its incidence is around 1500 new subjects per year in France. Among the numerous potential motor troubles of CP children, we will focus on facial dyspraxia. This dyspraxia that can be very severe for some children can bring several consequences on different levels. It is partly due to the difficulty that those children encounter to avoid coupling when controlling muscles close one to the others but also because this control is often done following an “all-or-nothing” scheme. In addition to the problems of oral communication, the troubles that can appear most usually are difficulties for feeding as it necessitates a good coordination of chewing and facial muscles. CP children have unfortunately a lack of such coordination.

Our approach is to develop a new rehabilitation technique based on the reproduction of movement to generate phonemes. To do this, we use the interaction with an avatar animated from the child’s own motions. Indeed, the classical technique for facial motion rehabilitation is to ask the child to reproduce a gesture previously performed by the therapist. This is based on the hypothesis that the child is able to compare his own motion and the therapist’s one. It thus admits that the child is able to convert these two motions into a similar representation in a common mental space. This hypothesis seems to be too strong for children having a too severe affection. The principle of the new rehabilitation tool we wish to develop in RePliCA project is to simplify this process of praxias comparison. It is based on the mirror neurons concept developed in the team of Rizzolatti and Fadiga. This theory stipulates that observing a motion activates neurons that are common with those that allow to perform the same motion. The principle of our tool is that during a rehabilitation session the child will observe simultaneously on the same screen an avatar, the virtual therapist’s one, performing the gesture to be done, and a second avatar animated from the motion he actually performs. To avoid the use of a too complex motion capture system, the child will be filmed by a simple video camera. A first technical challenge is thus to perform markerless facial motion capture.

A major challenge of the project deals with the ability of CP children to identify themselves with their avatar such that they can have the visual feedback needed for the exercise of motor control proposed by the therapist. We will thus study the influence of the different parameters needed to design the avatar (age, appearance, etc.) on this ability.

This new technique will be experimented during the RePLICA project on a group of CP children. A first evaluation criterion will be the attention and implication ability of the children within an interactive rehabilitation session. A connate goal is thus to design a new metric based of facial motion to provide him a feedback within this session and also to evaluate the progress of children along a rehab program.

Even if we focus on speech rehabilitation, the goal is indeed larger. We hypothesize that, if the rehab process we propose has positive impact, it will not only on speech production but also in other facial functions, and among them, feeding.