Deep Learning and Variational Analysis for High-Dimensional and Geometric Biomedical Data
MCML Authors
Vladimir Golkov
Dr.
Abstract
Vladimir Golkov
Dr.
Abstract
In this thesis, we use deep learning and variational analysis to solve various problems from biology and medicine related to advanced data structures. We predict the structure of proteins from their evolutionary statistics, and the function of proteins and small molecules from their structure. We also present image processing methods for diffusion MRI that reduce the scan duration by a factor of twelve and improve the image quality.
BibTeXKey: Gol21