Machine learning (and in particular deep-learning) methods have been at the centre of amazing progress in the field of computational image analysis. Much of this progress has centred around the analysis and classification of 2D images, with 3D applications being comparatively rare.
A central challenge in 3D microscopy is "tomographic reconstruction": the synthesis of a 3D image, from many 2D images of the sample. At first glance, this is a computational imaging challenge that involves large amounts of data, and thus an ideal application for machine-learning methods. However, tomographic reconstruction involves several unique computational and mathematical challenges, that have thus far frustrated the development of practical deep-learning and machine-learning algorithms.
In this project the student will work to develop machine-learning algorithms for tomographic reconstruction, and deploy these algorithms at the ANU CTLab imaging facility.
Mathematics (Fourier transforms, Multivariable calculus, Linear algebra)
Some experience in programming, preferably python.