Project Aims
The purpose of this project is to develop novel image analysis methods for automated analysis of breast tissue microarray cores. It will also investigate the extent to which these methods, when combined with methods for analysis of clinical outcome data, result in improved tools for clinical trials and discovery experiments. Improvements in terms of speed of analysis, repeatability and accuracy are expected to arise from this research. More specifically, the aims are:
- To develop novel computational methods for automatic localisation of subcellular compartments in images of breast tissue microarray cores.
- To develop automated scoring based on these methods and to validate against "manual" methods in which pathologists score the tissue microarrays subjectively.
- To investigate the extent to which automation of scoring helps survival analysis and discovery experiments using state-of-the-art Bayesian/artificial neural network methods with data from a large breast cancer trial.
Funded by the Breast Cancer Research Trust.
|
