Development of a Pre-Diagnosis Tool Based on Machine Learning Algorithms on the BHK Test to Improve the Diagnosis of Dysgraphia

Dysgraphia is a writing disorder that affects significant part of the population, especially school aged children and particularly boys. Nowadays, dysgraphia insufficiently diagnosed, partly because cumbersomeness existing tests. This study aims at developing an automated pre-diagnosis tool for allowing wide screening among children. Indeed, wider population would allow better care with handwriting deficits. based on world's largest known database samples uses supervised learning algorithms (Support Vector Machine). Four graphic tablets two acquisition software solutions were used, in order to ensure not tablet dependent can be used widely. A total 580 from 2nd 5th grade, which 122 dysgraphia, asked perform French version BHK test tablet. Almost hundred features developed these written tracks. The hyperparameters SVM most discriminating between without selected training dataset comprised 80% (461 children). With features, performances (119 children) sensitivity 91% specificity 81% detection dysgraphia. Thus, our has accuracy level similar human examiner. Moreover, it widely usable, its independence tablet, age thanks careful calibration use moving z-score calculation.