Dynamilis, une application pour l’apprentissage de l’écriture manuscrite - Thibault Asselborn

Notre matinée d’échanges scientifiques autour du thème des technologies favorisant l’apprentissage des personnes avec troubles du neuro-développement était composée de deux symposia:
Mieux lire et écrire grâce aux nouvelles technologies dans les TND - proposé par David Cohen (AP-HP)
Quelles caractéristiques des nouvelles technologies favorisent l’apprentissage des personnes avec TND? - Proposé par Mohamed Chetouani (Sorbonne, ISIR)

---------------------------------------------------------------------------------------------------------------
Découvrez l'intervention de Thibault Asselborn sur son application Dynamilis.

SYMPOSIUM 1 - INTERVENTION 3

Learning disabilities have a significant impact on children's lives. Reports show that up to 25% of children present learning and attention difficulties, which lead to loss of self-confidence, concentration difficulties, and fatigue. Our app, Dynamilis, focuses on a specific learning difficulty: handwriting. Even though we live in an increasingly digitized world, handwriting remains a critical skill to be acquired by every school child. Even with proper training, up to 25% of school children fail to fully master handwriting. Their difficulties cascade on all school activities that involve handwriting. Therefore, there is a correlation between handwriting problems, self esteem, and behavioral and academic development. We often meet some of these children who are not even 7 years old, but for whom school is already a very negative experience. Hence, it is of prime importance to detect handwriting difficulties as soon as possible in order to begin remediation.
We have developed Dynamilis, a software program running on iPad that analyzes the handwriting of pupils aged 5 to 12 and proposes activities according to the established handwriting profile.
Current tests are slow and often used too late in the child's curriculum. Our solution to analyze handwriting is a drastic improvement over the tests on paper that are currently in use to detect handwriting difficulties. Indeed, using a digital tablet allows us to extract not only static information (the shape of letters), but also the dynamics of handwriting (speed, pressure, acceleration, tilt, …). By processing these data using machine learning, we are able to establish handwriting profiles in 30 seconds (vs. 20 minutes for a therapist grading the current test), in a more objective way (current tests only have 60% inter-rater agreement), and to understand the nature of handwriting difficulties. Indeed, our solution is not restricted to giving a binary result (dysgraphic or non dysgraphic), but it computes a complete handwriting profile that identifies the specific strengths and weaknesses of a given child (scores with pressure, tilt, kinematic, and static aspects). This handwriting profile is then used to recommend remediation activities, co-created with more than 50 therapists, each of them focusing on one specific aspect of handwriting (control of the pressure, dexterity of fingers, precision, …).

---------------------------------------------------------------------------------------------------------------
Pour suivre nos actualités et rester en contact sur nos médias sociaux :
Facebook: https://www.facebook.com/profile.php?...
Twitter: https://twitter.com/iMIND_Lyon
Instagram: https://www.instagram.com/imind.lyon
Linkedin: https://www.linkedin.com/company/imin...
Site web: https://centre-imind.fr