Marie Schaer is an Associate Professor at the Department of Psychiatry at UNIGE with a specialisation in child psychiatry and neurodevelopmental disorders. After completing her medical studies, she obtained a PhD in neuroscience with a focus on brain imaging. Marie deepened her expertise in child psychiatry, receiving specialised training in brain imaging in autism at Stanford University. Since 2016, she has been in charge of the Geneva Autism Outpatient Clinic. There, she leads efforts in early detection and diagnosis.

Marie Schaer is fascinated by the understanding of the intricacies of brain development and how disturbances in this process contribute to neurological diversity. Her research aims to develop more effective interventions to support children and families struggling with atypical development, particularly autism spectrum disorder.

•  Page on the website of the Faculty of medicine

Developmental Trajectories in Young Children With Autism

Marie Schaer’s research team has been running the Geneva Autism Cohort project since its inception in 2012, as part of the NCCR Synapsy. This project focuses on following the development of a cohort of over 500 young children, mostly preschoolers (1–4 years old) with a confirmed diagnosis of ASD, as well as age-matched typically developing children. Using a range of neuroscience tools, including eye tracking, EEG, MRI and standardised tests, the team aims to identify predictive factors for future outcomes and gain insight into the impact of early intervention. The project’s main aim is to elucidate the variability in response to early intervention, with the aim of informing health policy and helping parents make decisions. The longitudinal nature of this project is important to understand the evolving needs and outcomes of children as they grow.

A Digital Phenotyping Strategy to Better Characterise Autism Spectrum Disorders

In the search for better predictors of future ASD outcomes, Marie Schaer recognises that current behavioural tools are limited in the granularity of describing each child’s profile. To achieve a deep phenotyping strategy in a seamless and scalable manner, her team proposes to rely on the development of computer vision and artificial intelligence to develop automated tools for fine-grained characterisation of autism spectrum disorders. Previous work from her lab, using video of children with and without autism interacting with an adult, has demonstrated that it is possible to use computer vision and pose estimation to provide a reliable diagnosis of autism from naturalistic recordings of social interactions. They have also worked on the prosody of vocal productions and found that certain features of the child’s voice are predictive of future language development. As they continue to improve their algorithms, the potential for large-scale screening is growing, promising a future of personalised therapeutic strategies and improved outcomes for children with ASD.