Improving brain health predictions by application of machine learning to genetic and connectomic data

Speakers: Gabriel Byczynski (Vanneste Lab) & Yihe Weng (Whelan Lab)

Abstract: The diagnosis of psychological disorders relies on the subjective, lived experience of the patient. However, symptoms and endophenotypes offer an objective measure of dysfunction, at a biologically more plausible level. The application of machine learning in brain data allows for large datasets of various data types, whether it be genetic or connectomic, to be leveraged for prediction and classification. This advancement leads to a deeper understanding of how different aspects of the brain can provide information about health, brain function, and disorders. In this talk, we will describe using machine learning methodologies in genetic and brain data to predict cognitive function and psychiatric symptoms. Specifically, Gabriel Byczynski (Vanneste Lab) will discuss the utilisation of genetic expression data to enhance the objective diagnosis of psychiatric symptoms. In addition, Yihe Weng (Whelan Lab) will present the application of machine learning to functional neuroimaging data to predict human cognitive processes and describe their associations with substance use.