Funding outcome

The objective of this research project is to develop an improved system for automatic analysis of 3D imaging data used to inform the treatment of young patients who require medical procedures to repair their hearts and the surrounding blood vessels. In the current state, doctors who are performing these procedures rely on manual visual inspection of X-ray images, but are typically unable to acquire quantitative measurements to inform their treatment. Especially in the context of time-sensitive procedures, there is the potential to have a positive impact on patient outcomes by providing additional information about the size, location, and structure of the heart and vessels of interest. This project will focus on the analysis of relevant X-ray imaging datasets that have been previously collected and anonymized from procedures at the Cardiac Catheterization Lab of the Mazankowski Alberta Heart Institute. The first stage of this project will focus on developing a reference dataset of X-ray images in which the specific structures of interest, the pulmonary artery and subsequent vessels, have been identified and digitally labelled. This will involve development of a robust labelling strategy that will require visual inspection by expert users to confirm accuracy. Once this labelled group of data has been generated from a sub-set of the total scan data acquired, it will be used to train a machine-learning algorithm which is capable of automatically labelling the pulmonary artery and subsequent vessels of interest. The performance of the generated algorithm will then be evaluated in terms of processing time and accuracy by performing automated segmentation on the remaining data sets.