Advanced image analysis for prostate cancer using functional imaging and histopathology

Project description

Prostate cancer is a disease where multiple tumour deposits are often seen in the prostate gland. In this Peter MacCallum Cancer Centre project, advanced image analysis and machine learning techniques will be developed and applied to co-registered histopathology and functional MR images to improve the understanding of prostate tumour location and biology. Overall this project has three aims:

Aim 1: Develop and apply machine learning methods to detect tumour location and tumour characteristics in histopathology specimens of the prostate.

Aim 2: Establish the relationship between tumour location and characteristics in histopathology with in vivo functional magnetic resonance (MR) imaging.

Aim 3: Apply machine learning techniques to extract 3D voxel maps from co-registered images for relevant radiobiological parameters, incorporating measures of uncertainty.

Ultimately this will provide a framework for designing improved radiotherapy treatment plans for patients with prostate cancer, to improve tumour control while reducing treatment side effects, by sparing surrounding tissues from high doses of radiation.

Project team

Leader: Hayley Reynolds

Staff: Annette Haworth, Scott Williams (Peter MacCallum Cancer Centre), Cheng Soon Ong (NICTA)

Collaborators: Matthew DiFranco (Medical University of Vienna), Nicholas Hardcastle (Peter MacCallum Cancer Centre), Gary Liney (Liverpool Hospital), Jason Dowling (CSIRO)

Sponsors: Peter MacCallum Cancer Centre

Other projects

Convergence of engineering and IT with the life sciences projects

Disciplines

Biomedical Engineering

Domains

Convergence of engineering and IT with the life sciences

Keywords

biomedical engineering; cancer; histopathology; magnetic resonance imaging MRI