FindAut is a startup based in Switzerland. We are developing an online tool
for the early detection of autism. FindAut ist part of the startup support programs at
Innosuisse - the Swiss Innovation Promotion Agency -
as well as be-advanced
of the Bernese government.
Today, autism spectrum disorder (ASD) is ususally diagnosed very late - on average at the age of 5 years (6.0 years in Great Britain or 4.4 in the US). However, scientific research indicates an early treatment is very important. An early intervention by the adequate therapy often has a very positive influence on the development of the ASD. The quality of life of patients and their families improves substantially. Symptoms of ASD are often detectable already at the age of 12 months. (see e.g. Bryson et al. 2004). Our tool aims at detecting signs of autism at the age of 18 months. Therewith improving the prospects of success for a professional therapy.
Long waiting times for a diagnosis are a widespread problem. Waiting times in Switzerland, for example, can amount up to one year. It is our goal to speed up and simplify the process of diagnosing autism. Our web application is intended to be used as a supportive tool in the course of diagnosis - after beeing certified according to applicable medical device regulations. To facilitate accessibility we aim for a web-based application, usable either with a smartphone or a desktop computer.
Recent Studies suggest a correlation between reactions to specific stimuli - like the showing of a picture - and the presence of an autism spectrum disorder (see e.g. Franchini et al. 2017 ; Woo Cho et al. 2016; Takarae et al. 2004; Shultz und Jones 2011). The most promising patterns of reactions relate to the eye focus and pupil movements (i.e. saccades). Our tool recognizes these patterns by applying a mathematical model, which uses methods of so called "Machine Learning". The analysis focuses on the saccades of the eyes, which are observed by the selfie camera of the smartphone or the desktop computer.Detecting eye movements:
The project is in an early development phase. FindAut ist part of the startup program of Innosuisse - the Swiss Innovation Promotion Agency. In November 2019, FinAut has been accepted to the "highflyer"-program of be-advanced of the Bernese government. Currently, we are developing the web application and are building partnerships with research institutions and associations.
Machine Learning needs a data base to detect patterns. Thereby, the raw data is saved as video recordings, while the effectively processed data points are deducted values such as relative pupil positions, eye movement speed or audio frequencies. Data protection is crucial for us, since we are processing health data in the broader sense as well as data of minors, which are particularily deserving protection.
We comply with the Swiss law and the General Data Protection Regulation (GDPR) of the European Union.
Christian Ruiz holds a PhD in quantitative social research (University of Lausanne). He specialized with a COS at the EPFL in Applied Machine Learning. Christian Ruiz has 10 years of experience in quantitative research and 5 years in the field of Machine Learning. He is father of a son, married and lives in Bern.
Alexander Hug lives in Zurich and has a background in quantitative social research aswell. He holds a master's degree from the IHEID in Geneva. Alexander Hug worked several years as a statistician at the Lucerne government and has experience in project management and as a founder of a consulting agency.
Hilel Dridi, who currently lives in Lausanne, is an IT professional with over 10 years of experience as an engineer, project manager and presales solution manager in France. He holds a COS from the EPFL in Applied Machine Learning and is father of an autistic kid.
Feel free to cantact us with your questions and inquiries at firstname.lastname@example.org
076 448 75 00