Project

Early Screening of Dyslexia Using a Language-Independent Game and Machine Learning

Goal: Our research aims to combine a playful approach with auditory and visual cues to predict the risk of having dyslexia.

Children with dyslexia have difficulties learning how to read and write. They are often diagnosed after they fail in school, even though dyslexia is not related to general intelligence. We present an approach for earlier screening of dyslexia using a language-independent game in combination with machine learning models trained with the interaction data. By earlier, we mean before children learn how to read and write.

To reach this goal, we designed the game content with the following areas in mind:
(1) Knowledge of the analysis of word errors from people with dyslexia in different languages; and (2) the parameters related to dyslexia and the auditory and visual perception. With our two designed games (MusVis and DGames), we collected data sets in different languages (mainly Spanish and German) and evaluated them with machine learning classifiers.
Our envisioned web application will contribute to 5% to 15% of the population by giving them a chance to succeed in life and find their skills to impress the world. Our results open the possibility of low-cost and early screening of dyslexia through the Web. Some of the results are already published, please look at our publication list.

The first preliminary study in Spanish, English, and German started.
Please contact me,
— If you would like to participate with your child, or
— If you would like to collaborate also in other languages.

Thanks, Maria

Methods: Human-Computer Interaction, Web Tools, Empirical Study, Quasi-Experiment

Date: 28 January 2016

Updates
0 new
19
Recommendations
0 new
3
Followers
0 new
40
Reads
2 new
586

Project log

Maria Rauschenberger
added a research item
Children with dyslexia have difficulties learning how to read and write. They are often diagnosed after they fail school even if dyslexia is not related to general intelligence. Early screening of dyslexia can prevent the negative side effects of late detection and enables early intervention. In this context, we present an approach for universal screening of dyslexia using machine learning models with data gathered from a web-based language-independent game. We designed the game content taking into consideration the analysis of mistakes of people with dyslexia in different languages and other parameters related to dyslexia like auditory perception as well as visual perception. We did a user study with 313 children (116 with dyslexia) and train predictive machine learning models with the collected data. Our method yields an accuracy of 0.74 for German and 0.69 for Spanish as well as a F1-score of 0.75 for German and 0.75 for Spanish, using Random Forests and Extra Trees, respectively. We also present the collected user data, game content design, potential new auditory input, and knowledge about the design approach for future research to explore universal screening of dyslexia. Universal screening with language-independent content can be used for the screening of pre-readers who do not have any language skills, facilitating a potential early intervention.
Maria Rauschenberger
added a research item
This is the user data that was collected with the game MusVis as well as analyzed and published in different publications. We designed the game content taking into consideration the analysis of mistakes of people with dyslexia in different languages and other parameters related to dyslexia like auditory perception as well as visual perception. We collected with the online experiment in total 313 participants. We separated our data into three data sets: one for the Spanish participants (ES, n = 153), a second for the German participants (DE, n = 149), and one for all languages (ALL, n = 313) in which we included participants that spoke English (n = 11). Participants ranged in age from 7 to 12 years old.
Maria Rauschenberger
added 3 research items
Protocol of the semi-structured literature review to select the content for 'A Universal Screening Tool for Dyslexiaby a Web-Game and Machine Learning'.
Maria Rauschenberger
added 2 research items
When discussing interpretable machine learning results, researchers need to compare them and check for reliability, especially for health-related data. The reason is the negative impact of wrong results on a person, such as in wrong prediction of cancer, incorrect assessment of the COVID-19 pandemic situation, or missing early screening of dyslexia. Often only small data exists for these complex interdisciplinary research projects. Hence, it is essential that this type of research understands different methodologies and mindsets such as the Design Science Methodology, Human-Centered Design or Data Science approaches to ensure interpretable and reliable results. Therefore, we present various recommendations and design considerations for experiments that help to avoid over-fitting and biased interpretation of results when having small imbalanced data related to health. We also present two very different use cases: early screening of dyslexia and event prediction in multiple sclerosis.
When discussing interpretable machine learning results, researchers need to compare results and reflect on reliable results, especially for health-related data. The reason is the negative impact of wrong results on a person, such as in missing early screening of dyslexia or wrong prediction of cancer. We present nine criteria that help avoiding over-fitting and biased interpretation of results when having small imbalanced data related to health. We present a use case of early screening of dyslexia with an imbalanced data set using machine learning classification to explain design decisions and discuss issues for further research.
Maria Rauschenberger
added a research item
Children with dyslexia are often diagnosed after they fail school even if dyslexia is not related to general intelligence. In this work, we present an approach for universal screening of dyslexia using machine learning models with data gathered from a web-based language-independent game. We designed the game content taking into consideration the analysis of mistakes of people with dyslexia in different languages and other parameters related to dyslexia like auditory perception as well as visual perception. We did a user study with 313 children (116 with dyslexia) and train predictive machine learning models with the collected data. Our method yields an accuracy of 0.74 for German and 0.69 for Spanish as well as an F1-score of 0.75 for German and 0.75 for Spanish, using Random Forests and Extra Trees, respectively. To the best of our knowledge, this is the first time that risk of dyslexia is screened using a language-independent content web-based game and machine-learning. Universal screening with language-independent content can be used for the screening of pre-readers who do not have any language skills, facilitating a potential early intervention.
Maria Rauschenberger
added a research item
Children with dyslexia have difficulties learning how to read and write. They are often diagnosed after they fail in school, even though dyslexia is not related to general intelligence. In this thesis, we present an approach for earlier screening of dyslexia using a language-independent game in combination with machine learning models trained with the interaction data. By earlier, we mean before children learn how to read and write. To reach this goal, we designed the game content with knowl- edge of the analysis of word errors from people with dyslexia in different languages and the parameters reported to be related to dyslexia, such as auditory and visual perception. With our two de- signed games (MusVis and DGames), we collected data sets (313 and 137 participants) in different languages (mainly Spanish and German) and evaluated them with machine learning classifiers. For MusVis we mainly use content that refers to one single acoustic or visual indicator, while DGames content refers to generic content related to various indicators. Our method provides an accuracy of 0.74 for German and 0.69 for Spanish and F1-scores of 0.75 for German and 0.75 for Spanish in MusVis when Random Forest and Extra Trees are used. DGames was mainly evaluated with German and reached a peak accuracy of 0.67 and a peak F1-score of 0.74. Our results open the possibility of low-cost and early screening of dyslexia through the Web.
Maria Rauschenberger
added a research item
Dyslexia is a widespread specific learning disorder, which can have a particularly negative influence on the learning success of children. Early detection of dyslexia is the foundation for early intervention, which is the key to reduce the adverse effects of dyslexia, e.g., bad school grades. In this paper, we present the prototype of a puzzle app, which we explicitly designed with the human-centred design (HCD) process for dyslexia screening in pre-reader using new indicators related to motor skills to ensure users needs for the data collection to apply machine learning prediction. The app records the telemetry of the gaming sequence in order to derive future statements about the prevalence of dyslexia based on the telemetry data. The high-fidelity prototype was evaluated with a five-user test usability study with five German-speaking child-parent pairs. The results show how young children and parents are interacting with new games, and how new applications (web and mobile technologies) which are used for online experiments, could be developed. The usability of the prototype is suitable for the target group with only minor limitations. CCS
Maria Rauschenberger
added an update
Hi everyone,
We made a huge effort and collected, clustered, and compared different screen web applications for dyslexia. We present the result in our chapter "Technologies for Dyslexia" available in the book Web Accessibility.
Our chapter includes past, current, and future research technologies to support, intervene, or detect people with dyslexia!
The other chapters are also a great resource on Web Accessibility! If you are looking for details or a starting point for your research as well as detailed information on a specific topic like "Mathematics and Statistics", "Multimedia Accessibility", "Wearables", "Mobile Web“... It is your book about recent Web Accessibility development.
Thanks to my co-authors Ricardo Baeza-Yates and Private Profile
as well as the editors Yeliz Yesilada and Simon Harper
and do not forget to check out the other chapters !!!
 
Maria Rauschenberger
added a research item
Nowadays, being excluded from the web is a huge disadvantage. People with dyslexia have, despite their general intelligence, difficulties for reading and writing through their whole life. Therefore, web technologies can help people with dyslexia to improve their reading and writing experience on the web. This chapter introduces the main technologies and many examples of tools that support a person with dyslexia in processing information on the web, either in assistive applications for reading and writing as well as using web applications/games for dyslexia screening and intervention.
Maria Rauschenberger
added a research item
Using serious games to screen dyslexia has been a suc- cessful approach for English, German and Spanish. In a pilot study with a desktop game, we addressed pre-readers screening, that is, younger children who have not acquired reading or writing skills. Based on our results, we have redesigned the game content and new interactions with visual and musical cues. Hence, here we present a tablet game, DGames, which has the potential to predict dyslexia in pre-readers. This could contribute to around 10% of the population that is affected by dyslexia, as children will gain more time to learn to cope with the challenges of learning how to read and write.
Maria Rauschenberger
added a research item
Detecting dyslexia is important because early intervention is key to avoid the negative effects of dyslexia such as school failure. Most of the current approaches to detect dyslexia require expensive personnel (i.e. psychologists) or special hardware (i.e. eye trackers or MRI machines). Also, most of the methods can only be used when children are learning how to read but not before, necessarily delaying needed early intervention. In this work, we present a study with 178 participants speaking different languages (Spanish, German, English, and Catalan) with and without dyslexia using a web-based game built with musical and visual elements that are language independent. The study reveals eighth game measures with significant differences for Spanish children with and without dyslexia, which could be used in future work as a basis for language independent detection. A web- based application like this could have a major impact on children all over the world by easily screening them and suggest the help they need.
Maria Rauschenberger
added a research item
An initial version of this poster was presented at W4A 2017 , Perth, Western Australia: M. Rauschenberger, L. Rello, R. Baeza-Yates, E. Gomez, and J. P. Bigham, Towards the Prediction of Dyslexia by a Web-based Game with Musical Elements. In W4A’17: International Web for All Conference, 2017, pp. 4–7. http://doi.org/10.1145/3058555.3058565
Maria Rauschenberger
added an update
Once a year the German Reading Award (Deutsche Lesepreis) is presented in Berlin by the Foundation Lesen and the Foundation Commerzbank-Stiftung. Projects in 4 different categories are honored for their approach, engagement or make which have a beneficial effect on reading. Maria Rauschenberger is honored for her Ph.D. idea and pilot study on a playful and early approach to screen children with dyslexia while playing a game.
 
Maria Rauschenberger
added an update
Our Game and Approach are nominated for the German Reading Award “Deutscher Lesepreis 2017”. Out of 281 are we under the last 6 projects from all over Germany in the category ideas for tomorrow “Ideen für morgen”.
The foundation Stiftung Lesen und Commerzbank-Stiftung is presenting the award. http://bit.ly/LesepreisNominierungen
 
Maria Rauschenberger
added an update
The school authorities of Lower Saxony Germany confirmed and permitted our study approach for MusVis!
 
Maria Rauschenberger
added an update
The department of education of the north region German named Schleswig-Holstein confirmed our study approach for MusVis and the first school already participated! Thanks to all children, parents and teachers for helping us to conduct the study!
 
Maria Rauschenberger
added an update
We have done a lot in the last weeks. Updates on the project hopefully soon with new research content! Have a good weekend research fellows!
 
Maria Rauschenberger
added an update
The first phase of the study has started. We are looking for collaborators to translate our application into different languages and to conduct a research study. If you are interested please let us know and get in contact with us.
 
Maria Rauschenberger
added an update
Private Profile and Ricardo Baeza-Yates present the poster at the Web for All (W4A) Conference in Perth, Australia.
Also, the the video of the paper presentation from Luz is online: http://bit.ly/2oUWmZa
 
Maria Rauschenberger
added 2 research items
Paper Presented at the Web for All Conference in Perth, Australia M. Rauschenberger, L. Rello, R. Baeza-yates, E. Gomez, and J. P. Bigham, “Towards the Prediction of Dyslexia by a Web-based Game with Musical Elements,” w4a, pp. 4–7, 2017. The generated musical elements are available at http://bit.ly/2jeejmC The demo of the prototype DysMusic is available at http://bit.ly/DysMus More Project can be found on www.mariarauschenberger.com Abstract: Current tools for screening dyslexia use linguistic elements,since most dyslexia manifestations are related to difficultiesin reading and writing. These tools can only be used withchildren that have already acquired some reading skills and; sometimes, this detection comes too late to apply proper re-mediation. In this paper, we propose a method and presentDysMusic, a prototype which aims to predict risk of hav-ing dyslexia before acquiring reading skills. The prototypewas designed with the help of five children and five parentswho tested the game using the think aloud protocol and being observed while playing. The advantages ofDysMusicare that the approach is language independent and could beused with younger children,i.e., pre-readers.
Maria Rauschenberger
added an update
Hi everyone -
just uploaded
  • the Authors Version of the Paper: Towards the Prediction of Dyslexia by a Web-based Game with Musical Elements were we present our Usability Study of the Prototype DysMusic. DysMusic is the Musical Part of our Application MusVis.
  • The Musical Element, used for the Prototype Usability Study, are available.
  • Video of the interaction of DysMusic
Just look at the paper on Researchgate and the uploaded Resources!
 
Maria Rauschenberger
added a research item
Current tools for screening dyslexia use linguistic elements, since most dyslexia manifestations are related to difficulties in reading and writing. These tools can only be used with children that have already acquired some reading skills and; sometimes, this detection comes too late to apply proper remediation. In this paper, we propose a method and present DysMusic, a prototype which aims to predict risk of having dyslexia before acquiring reading skills. The prototype was designed with the help of five children and five parents who tested the game using the think aloud protocol and being observed while playing. The advantages of DysMusic are that the approach is language independent and could be used with younger children, i.e., pre-readers.
Maria Rauschenberger
added an update
The prototype of the musical part was designed with the help of five children and five parents who tested the game using the think aloud protocol and being observed while playing. The work will be presented at Web for All (w4a) in Australia. Next is the pilot study for different languages.
Do you want to participate?
Let's play - participate in our research
 
Maria Rauschenberger
added a research item
The aim of this research is to show that a playful approach combined with music can detect children with dyslexia. Early detection will prevent children from suffering in school until they are detected due to bad grades. Our envisioned web application will contribute to 10% of the population by giving them a chance to succeed in life and find their skills to impress the world.
Maria Rauschenberger
added an update
LAUNCH OF THE STUDY
We launched the study for all three languages.
Participants requirements:
  • speak either German, Spanish or English
  • Children at the age of 7 till 11
Further Information at http://bit.ly/2kMcqi7
 
Maria Rauschenberger
added an update
We are updating the application on visual elements - and are on the way to launching the study.
If you like to participate please send me a message.
Participants need to speak either English, German or Spanish:
Need to be 4 till 11 year old and have access to the internet via a tablet ( e.g. iPad)
 
Maria Rauschenberger
added an update
Internal Test: Front-end and Back-end are in the internal test phase. After that: If you would like to participate we are now collecting the first people for the try out. Just send me a message.
 
Maria Rauschenberger
added an update
Good news - the fronted of the application is in the test phase!
Attached the "Face of the application" DysMusic!
Next the backend programming will be finished!
 
Maria Rauschenberger
added an update
UPF Magazine - Music fight Dyslexia - https://issuu.com/universitatpompeufabra/docs/num_12_eng
We are still working on the implementation of the application. Hopefully we have news soon!
 
Maria Rauschenberger
added 14 project references
Maria Rauschenberger
added a project goal
Our research aims to combine a playful approach with auditory and visual cues to predict the risk of having dyslexia.
Children with dyslexia have difficulties learning how to read and write. They are often diagnosed after they fail in school, even though dyslexia is not related to general intelligence. We present an approach for earlier screening of dyslexia using a language-independent game in combination with machine learning models trained with the interaction data. By earlier, we mean before children learn how to read and write.
To reach this goal, we designed the game content with the following areas in mind:
(1) Knowledge of the analysis of word errors from people with dyslexia in different languages; and (2) the parameters related to dyslexia and the auditory and visual perception. With our two designed games (MusVis and DGames), we collected data sets in different languages (mainly Spanish and German) and evaluated them with machine learning classifiers.
Our envisioned web application will contribute to 5% to 15% of the population by giving them a chance to succeed in life and find their skills to impress the world. Our results open the possibility of low-cost and early screening of dyslexia through the Web. Some of the results are already published, please look at our publication list.
The first preliminary study in Spanish, English, and German started.
Please contact me,
— If you would like to participate with your child, or
— If you would like to collaborate also in other languages.
Thanks, Maria