Automatizovaná diagnóza vývojové dysgrafie založená na kvantitativní analýze online písma

Investor logo

Warning

This publication doesn't include Faculty of Economics and Administration. It includes Faculty of Arts. Official publication website can be found on muni.cz.
Title in English Diagnosis of Developmental Dysgraphia Based on Quantitative Analysis of Online Handwriting
Authors

VOJTĚCH Zvončák ŠAFÁROVÁ Katarína MEKYSKA Jiří MUCHA Ján KISKA Tomáš LOSENICKÁ Barbora ČECHOVÁ Barbora FRANCOVÁ Pavlína SMÉKAL Zdeněk

Year of publication 2018
Type Article in Periodical
Magazine / Source Elektrorevue
MU Faculty or unit

Faculty of Arts

Citation
Web http://www.elektrorevue.cz/cz/clanky/zpracovani-signalu/0/automatizovana-diagnoza-vyvojove-dysgrafie-zalozena-na-kvantitativni-analyze-online-pisma-1/
Keywords developmental dysgraphia; children dysgraphia; digitizing tablet; HPSQ; random forests; support vector machine
Description The prevalence of handwriting difficulties among school-aged children is around 10–30 %. Until now, there is no objective method to diagnose and rate developmental dysgraphia (DD) in Czech Republic. The goal of this study is to propose a new method of objective DD diagnosis based on quantitative analysis of online handwriting. For this purpose, we extracted a set of spatial, temporal, kinematic and dynamic features from three handwriting tasks. Consequently, we performed a correlation analysis between these features and score of handwriting proficiency screening questionnaire (HPSQ), in order to identify parameters with a good discrimination power. Using random forests classifier in combination with quantification of alphabet writing task, we reached nearly 80% classification accuracy (77% sensitivity, 83% specificity). The classification accuracy was increased to 92% (92% sensitivity, 93% specificity) by employing SFFS (Sequential Forward Feature Selection) method. This pilot study proves the possibility of automatic DD diagnosis in children cohort writing with cursive letters.
Related projects:

You are running an old browser version. We recommend updating your browser to its latest version.