When segmenting medical image data, have you ever struggled with separating two bones from each other? Well, you’re not the only one.
Hannelore Boey is conducting her PhD research at the Biomechanics group of the University of Leuven. She is interested in studying how individual foot bones move with respect to each other during motion. In order to do this, she works with 4D CT scans. Regular CT scans show a static image, but 4D CT scanners are capable of taking multiple CT images per second. In her experiments, Hannelore has already collected a large number of datasets of 4D CT scans of the foot and ankle. The first necessary step for performing the measurements is that all the 28 (!) different bones in the foot need to be segmented and separated from each other in at least one of the frames. A daunting task!
To add to the complexity, the 4D CT data is often of poor quality, which after thresholding, renders the region growing tool unusable for separating individual bones from the mask (see illustrations). She had to resort to using the multiple slice edit tool to manually separate two adjacent bone masks. Therefore, it took more than one hour to segment the whole foot.
Luckily, in Mimics 19, she found a tool that could help her speed up the work. The new 'Split Mask' tool allows the separation of two masks in only a couple of seconds instead of a couple of minutes. Just imagine the time gained when separating all 28 bone masks! The time spent on medical image segmentation and the frustration she experienced when two masks did not want to split was greatly reduced. Hannelore said “This tool is a great improvement to my segmentation workflow, and is going to save me a lot of time over the course of my PhD!”
Do you recognize these obstacles?