FEIT Research Project Database

Microscopy image segmentation using machine and deep learning

Project Leader: Vijayaraghavan Rajagopal
Collaborators: Eric Hanssen (Bio21 Institute)
Primary Contact: Vijay Rajagopal (vijay.rajagopal@unimelb.edu.au)
Keywords: cancer; cardiovascular disease; computer vision; deep learning
Disciplines: Biomedical Engineering,Computing and Information Systems

The Cell Systems and Mechanobiology Group is a leader in the application of engineering methods to understand and discover new biology. We are particularly passionate about visualising, understanding and predicting how cells work and respond to their environment. 

The biomedical sciences have recently seen an explosion in microscopy imaging data thanks to revolutionary technologies like 3D serial block-face electron microscopy, light sheet microscopy and super resolution microscopy. These datasets provided unprecedented views of the architecture of cell systems but require significant amounts of time to manually annotate before new discoveries can be made. 

We are seeking outstanding engineering students with an interest in computer vision and deep learning to join our team to advance microscopy image segmentation using high throughput methods like deep learning convolutional neural networks (CNN). Successful applicants will customise existing CNNs to applications of interest and also make innovations in deep learning. We have on-going research projects in studying cardiac biology and cancer biology. While students will be focussed on one specific project for their PhD they will also have opportunities to participate in other active projects within the Cell Systems and Mechanobiology group using their capabilities and tools they develop.

Further information: https://biomedical.eng.unimelb.edu.au/cell-systems-mechanobiology/ https://www.sciencedirect.com/science/article/pii/S1047847718300595 https://www.biorxiv.org/content/biorxiv/early/2020/07/15/2020.07.15.203836.full.pdf https://bmcmedinformdecismak.biomedcentral.com/articles/10.1186/s12911-019-0962-1