Open Peer Review on Qeios Parents’ mHealth App for promoting dyslexia biomarker detection in children at home or at school: Feasibility, Acceptability, Economic impact, Pilot Study and Survey Results Gunet Eroglu, Mertali Köprülü 1 , Berdan Karabacak 2 1 Isik University 2 Neuro Brain Academy Funding: No specific funding was received for this work. Potential competing interests: No potential competing interests to declare. Abstract The use of mobile apps in diagnosing and improving health conditions has increased in recent years. If dyslexia is diagnosed at 7 years old, dyslexic children’s lives should evolve in a better way. Therefore, dyslexia diagnosis at an early age is very important but mostly delayed to give a chance to overcome any maturation delays. In the literature, there is no clinically tested mobile app for dyslexia biomarker detection that can objectively monitor the child’s situation at home or at school. No other research assessed the feasibility and acceptability of using this kind of mobile app to detect dyslexia with a survey on dyslexic families. Many different dyslexia biomarker detection methods exist in the literature. These methods are based on questionnaires, MRI scans, PET scans, EEG scans, and eye-tracking scans using Machine Learning methods. Each of these methods has its own drawbacks, such as the psychometric tests taking more than 1 hour; MRI scans and eye tracking solutions being expensive and being difficult to collect data, and the results may not be accurate enough to generalize as dyslexia have many subtypes. Here we present a novel mobile app that has an embedded dyslexia biomarker based on Z-scored QEEG data that has accomplished a high accuracy rate in diagnosing dyslexia. The mobile app can be used at home by parents or teachers at school. We have collected data from 207 children (96 of them have dyslexia, 111 of them are typically developing) between 7-10 years old for 60 sessions during their neurofeedback sessions. The data consists of the eyes-open resting state 2-minute QEEG data from 14-channels. In order to standardize the data, the Z-scores are calculated. Using the ANN machine learning method, dyslexic/normal classification has been done with a high accuracy rate (98.8%). ANN yields the highest accuracy results with standardized QEEG data in the literature. Auto Train Brain is a novel mobile app that has dyslexia biomarker detection software embedded into it. A survey is created to assess the mobile app’s dyslexia biomarker detection module’s feasibility, acceptability, and economic impact. The results have shown that the app module is found feasible and acceptable by families, however, it is found expensive to use at home. Günet Eroğlu a , Mertali Köprülü b , Berdan Karabacak c a Computer Engineering Dept, Engineering and Nature Faculty, Bahçeşehir University, Istanbul, Turkey Qeios, CC-BY 4.0 · Article, October 28, 2022 Qeios ID: 4W9RXU.3 · https://doi.org/10.32388/4W9RXU.3 1/17