Graduate Seminar Series: Clinical Stream
Graduate Seminar Series for the Institute of Biomedical Engineering (BME). This day is for clinical stream presenters.
If you would like to invite your Principal Investigator, please add their email via the ‘Add Guest’ button and they will also be notified of your presentation.
Presentation Title: Challenging the Labels of Autism and ADHD – A Data-Driven Graph-Based Approach
More than 600,000 children in Canada are diagnosed with autism spectrum disorder (ASD) and attention-deficit/hyperactivity disorder (ADHD). These diagnoses are currently based on specific behaviours (e.g., social communication difficulties, inattention), which may not correspond to the distinct and unique underlying biology. Moreover, the labels of ASD and ADHD often do not predict a child’s response to a specific treatment. Instead, care decisions are made based on each child’s unique behavioural presentation. In many cases, existing treatments are not effective, can have adverse side-effects, and many children continue to experience significant distress and often have poor outcomes as adults (e.g., mental health, employment, and independence). Collectively, these challenges suggest that ASD and ADHD may not exist as uniquely-defined diagnostic constructs and highlight the need to discover other groupings that may be more closely aligned with biology and/or response to treatment. To address this need, the proposed project will perform data-driven approach to question the validity of current diagnostic labels and identify new groupings for children with ASD and ADHD that are biologically relevant. I will do this through an interdisciplinary approach that employs computer science-based machine learning approaches to go beyond traditional analytics methods in this field. The objective of this project is to validate existing groupings or discover new groupings based on unique and distinct biological data. To this
end, I will pursue a novel analytical approach: instead of comparing diagnostic groups as currently done, I use an innovative data-driven approach that looks to the data to discover new groups that transcend the existing diagnostic labels or those that validate the current groups but that also explain why the current treatments are ineffective for many individuals in part because of their adverse side effects. I will do this by use of graph clustering method and graph neural network. This approach is capable of identifying groups who share similar neuroanatomical characteristics/complexities, regardless of diagnosis.
Supervisor Name: Dr.Azadeh Kushki
Year of Study: 4
Program of Study: PhD
Zoom link: https://us02web.zoom.us/j/89610372821?pwd=azd4SCtYVWtreVovaGNPV1c2NGY2Zz09
Meeting ID: 896 1037 2821
Powered by Calendly.com