{"title":"Handwriting treatment and acquisition in dysgraphic children using a humanoid robot-assistant","authors":"Soukaina Gouraguine, Mohammed Qbadou, K. Mansouri","doi":"10.1109/EDUCON52537.2022.9766701","DOIUrl":null,"url":null,"abstract":"Handwriting is one of the basic skills and an important means of communicating with others and expressing oneself, as it is important in all aspects of life. Acquiring handwriting is also a complex skill that takes years of training to be mastered. Advances in technology have allowed robots to accomplish a diversity of human activities, provoking interest from educators, researchers, and practitioners in discovering the potential advantages of employing robots as an intervention for children with dysgraphia who have difficulty automating their handwriting. Dysgraphia can be treated very well thanks to a writing rehabilitation adapted to the needs and the importance of the child’s problem. The purpose of this article is to explore the potential benefits of integrating a social humanoid robot in interventions for the treatment of dysgraphia in children. In this context, our research focuses on introducing a social humanoid robot assistant in an educative context to assist dysgraphic children to acquire handwriting. This work is done by applying a new approach based on a deep learning classification algorithm using convolutional neural networks (CNN) to determine the presence of dysgraphia from the handwriting of elementary school children. In this study, we realized a humanoid robot that will assist a teacher by equipping the NAO robot: (1) to moderate a session of learning to handwrite an alphabet character, (2) to detect whether a child is dysgraphic or not using a convolutional neural network, and (3) to assist and monitor dysgraphic children by performing tasks and suggesting rehabilitation sessions. The results indicate that it can distinguish dysgraphic children from non-dysgraphic children with an accuracy of 75%, a specificity of 75%, and a precision of 60%. The results reveal that the robot was able to classify learners so that the human tutor could assign the students to the appropriate rehabilitation program based on their specific needs.","PeriodicalId":416694,"journal":{"name":"2022 IEEE Global Engineering Education Conference (EDUCON)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2022 IEEE Global Engineering Education Conference (EDUCON)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/EDUCON52537.2022.9766701","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Handwriting is one of the basic skills and an important means of communicating with others and expressing oneself, as it is important in all aspects of life. Acquiring handwriting is also a complex skill that takes years of training to be mastered. Advances in technology have allowed robots to accomplish a diversity of human activities, provoking interest from educators, researchers, and practitioners in discovering the potential advantages of employing robots as an intervention for children with dysgraphia who have difficulty automating their handwriting. Dysgraphia can be treated very well thanks to a writing rehabilitation adapted to the needs and the importance of the child’s problem. The purpose of this article is to explore the potential benefits of integrating a social humanoid robot in interventions for the treatment of dysgraphia in children. In this context, our research focuses on introducing a social humanoid robot assistant in an educative context to assist dysgraphic children to acquire handwriting. This work is done by applying a new approach based on a deep learning classification algorithm using convolutional neural networks (CNN) to determine the presence of dysgraphia from the handwriting of elementary school children. In this study, we realized a humanoid robot that will assist a teacher by equipping the NAO robot: (1) to moderate a session of learning to handwrite an alphabet character, (2) to detect whether a child is dysgraphic or not using a convolutional neural network, and (3) to assist and monitor dysgraphic children by performing tasks and suggesting rehabilitation sessions. The results indicate that it can distinguish dysgraphic children from non-dysgraphic children with an accuracy of 75%, a specificity of 75%, and a precision of 60%. The results reveal that the robot was able to classify learners so that the human tutor could assign the students to the appropriate rehabilitation program based on their specific needs.