Scaleup Lexplore has won the first prize in the category Digital Wellbeing of the EIT Digital Challenge 2017. The Swedish tech venture has developed a method using artificial intelligence and machine learning to identify children at risk of dyslexia. By using Lexplore’s solution, schools can detect children with reading and writing problems early on. We interviewed Fredrik Wetterhall, CEO and co-founder of Lexplore, on winning the prestigious award and their plans for the future.
A major problem
According to Dominik Krabbe, EIT Digital Challenge Lead, dyslexia is a major problem in the world. Krabbe: “Dyslexia is a difficult word for a pretty common problem that can make life tough: reading and writing disabilities. It is estimated that between 10 and 20 percent of the population suffers from dyslexia. This amount may vary by country.” And up to this point, it has been really time consuming for schools to identify a child with dyslexia and subsequently giving them the proper help they need.
A high rate of effectiveness
This is where tech venture Lexplore comes in. As stated earlier, the company has developed an AI-based solution to identify children with reading and writing problems early on. The solution is also very accurate, as it has a 95% rate of effectiveness. Because of this, schools can use their resources more effectively and focus on the children that need extra help and support.
The problems of using traditional methods
According to Lexplore, their method enables schools to screen all of their students in a resource-efficient and accurate manner. Fredrik Wetterhall, CEO and co-founder of Lexplore, had the following to say, regarding the more traditional approaches to detecting children at risk of dyslexia: “We started Lexplore, because we saw that too many children with reading and writing problems were discovered way too late. The reason for this is because traditional methods of identifying children at risk of dyslexia require a lot of time consuming and manual work. For example, schools usually only test individual children whose teachers and parents suspected they may have dyslexia. Also, children’s results are hard to compare since this kind of data is quite subjective, dependent on who has done the test. We wanted to solve these problems for the pupils and the schools. The method was developed at the Karolinska Institute and initial research was funded by the Swedish government and multiple private investors.”
Wetterhall: “By using our solution, schools can save a lot of time. The test is carried out at schools with the help of the teachers, and only takes a couple of minutes to complete. Most of the time, children don’t even perceive what they are doing for us as a test. They will be asked to read two short texts on a computer screen, which is equipped with an eye tracker. After they have read the texts, the test is already over. And that’s it. Children with reading and writing problems will be identified early on and will experience a lot less of the negative effects that are associated with it.”
The importance of an early intervention
It has been proven that an early intervention achieves better results than one started at a later stage. Wetterhall: “When people are still young, it’s easy to learn to write and read. Also, if a child’s reading and writing problems are discovered at a later stage, he or she may have fallen behind too much to be able to catch up at school. It can also have a negative effect on the self-esteem of the child. This would be the result of not knowing why you’re having so much trouble with reading and writing.” Therefore, an early intervention is vital to give the concerned children the help they need.
The data behind the solution
The CEO and co-founder also explained how the solution works. Wetterhall: “I’d like to emphasize that we don’t actually diagnose dyslexia. Rather, we identify children who are at risk of having it and children who aren’t. This identification is made possible by a huge amount of empirical data we have collected since 1991.” This pile of data is the result of studies that surveyed the development of the reading ability of certain individuals from childhood to adulthood. According to Lexplore, it has been established in these studies that there is a link between eye movement and reading ability. Eye movements during reading would reflect the underlying problem with decoding words that affects people with dyslexia. The scientific results of Lexplore’s method were presented in the article “Screening for Dyslexia Using Eye Tracking During Reading”.
EIT Digital Challenge 2017
Since Lexplore won the Digital Wellbeing category of the Eit Digital Challenge 2017, the company will receive the first prize, including a cash prize of €50.000 and a year of full support during the EIT Digital Accelerator worth an additional €50.000. “Thanks to us participating in the EIT Digital Challenge, we have a lot of new contacts and it was rewarding to learn about the other great companies participating in the challenge. And this is only the starting point: the big value will be next year when we will be part of the EIT Digital Accelerator. As a result, we will gain a lot more visibility and contacts within EIT Digital and its partners.” This was stated by Wetterhall in an interview with EIT Digital.
Lexplore has more plans for the future though. Wetterhall: “At the moment, our business is active in Sweden and the United States. Next year, we will be expanding to the United Kingdom as well. We already have plans for expanding to Germany and the Netherlands, and EIT Digital will be a great help during our analysis of these markets. We hope to be able to reach these markets by 2019.”