Mariam Zabihi & Gonzalo Castro Leal - CMIC/WEISS Joint Seminar Series

Speaker: Mariam Zabihi

Title:  Finding a needle in a haystack: The Challenges of Identifying Biomarkers of Psychiatric Disorders

Abstract: 

Psychiatric disorders are all highly heterogeneous, both in terms of clinical presentation and underlying biology. Despite considerable effort towards understanding psychiatric disorders using neuroimaging biomarkers, heterogeneity remains a barrier, partly because studies mostly employ case-control approaches, which assume that the clinical group is homogeneous. Here, I describe the normative modelling approach as an alternative approach to the case-control paradigm to parse heterogeneity in Autism as an example of a highly heterogeneous psychiatric disorder. Next, I’ll talk about  optimal representation of neuroimaging data for understanding individual differences in brain systems and how these relate to behaviour.

Speaker: Gonzalo Castro Leal

Title: Crop Filling: artefact removal for real-world clinical neuroimaging data

Abstract: 

This talk is split into two parts, first I'll give a brief overview of my research background in medical image analysis. The main part of the talk will focus on the project I am involved; using clinical data to train EBM models for Alzheimer’s disease. Currently, the main challenge I am facing is solving the artefact that was found in the clinical dataset for this project, which is the partial brain coverage in T1w MRI scans. The current approach involves using super resolution methods in the T2w scans from the patients to fill in the missing data in the T1w scan. I will present the methodology and validation process that I am following and some of the results I have got so far.

Chair: James Cole