A deep generative model approach to the genetic analysis of medical images
I will introduce a new deep generative model for the genetic analysis of medical images, which combines convolutional neural networks and structured linear mixed models  to extract latent imaging features in the context of genetic association studies. I will present an application of the method to brain MRI images from the Alzheimer's Disease Neuroimaging Initiative dataset, where we reveal novel and known risk genes for neurological and psychiatric disorders.  Casale FP, Dalca A, Saglietti L, Listgarten J, Fusi N. Gaussian Process Prior Variational Autoencoders. InAdvances in Neural Information Processing Systems 2018 (pp. 10369-10380).
Francesco Paolo Casale is a data scientist at Insitro, a drug discovery company which aims to develop a new approach to drug development by combining large-scale data generation and machine learning tools. Previously, he was a postdoctoral researcher at Microsoft Research New England, where he worked on methods for automated machine learning and on deep learning models for imagining genetics. Before that, Paolo did a PhD in statistical genetics with Oliver Stegle at the EMBL-European Bioinformatics Institute and the University of Cambridge. He holds a bachelor’s and master’s degree in physics from the University of Naples, Italy.