Description: The Wall Lab is looking for a PhD-level data scientist with expertise in machine learning approaches to work with our mobilized phenotype group in developing technologies and methods to better detect, track, and treat brain and developmental health concerns, with a focus on autism spectrum disorder.
We have developed a therapeutic computer vision system for tracking emotions in faces that provides real-time social cues to the wearer, a child with autism. Referred to as the Autism Glass Project, this project is in the midst of a clinical trial aimed at evaluating the efficacy of the device when it is used over a period of time in the home independently of behavioral interventionists. This position offers the opportunity to be intimately involved in enhancing this technology; for instance, in integrating new computer vision and similar sensory AI tools. Taking a more human-computer interaction-oriented perspective, we will also seek to explore new ways in which such a tool can be used.
We have also launched a program to mobilize diagnostic screening and longitudinal measurement of improvements while children with autism and other developmental conditions engage in social behavioral therapy at home. Efforts of this form have taken several angles, from using machine learning techniques to refine and abbreviate diagnostic screeners, to gathering and analyzing both genome and microbiome data on large groups of participants, to a crowdsourcing campaign aimed at creating a community of families of those with autism and care providers. Key work in the future involves refinements of diagnostic tools, analyses of this growing pool of data using novel learning techniques, and scaling up all efforts, with the intent of building a community that is a resource for those with autism, scientists working on novel diagnostics and treatments, and clinicians.
These approaches fit together to define a major thrust of the Wall Lab - the mobilized phenotype group for detection and treatment technologies. All work is mean to have the possibility for strong translational impact.
Expertise should include techniques in computer vision and development and use of supervised and unsupervised machine learning algorithms. The ideal candidate has experience in deep learning, including convolutional and recurrent neural networks, as well as integrating such technologies on mobile platforms. Emphasis is put on being able to make working prototypes with the help of other developers. The mobilized phenotype group works on projects largely in teams, and as such, previous team-based and leadership experiences are highly valued.
Application Instructions: Please email the poster, including a CV.