Liver Segmentation from CT Images for Surgery
Author Information
Author(s): Dário AB Oliveira, Raul Q Feitosa, Mauro M Correia
Primary Institution: Pontifical Catholic University of Rio de Janeiro
Hypothesis
This study proposes a methodology to segment the liver, its vessels, and nodules from computer tomography images for surgical planning.
Conclusion
The method for liver segmentation performed well, achieving consistent results with the anatomy of the liver.
Supporting Evidence
- The method was ranked as the 6th best semi-automatic method in a liver segmentation competition.
- Good performance was achieved in 17 out of 20 exams.
- The segmentation results were visually consistent with liver anatomy.
Takeaway
The researchers created a way to help doctors see the liver and its parts better in CT scans, which can make surgery easier.
Methodology
The methodology consists of four steps: liver segmentation using level sets, segmentation of vessels and nodules using a Gaussian mixture model, identification of hepatic and portal veins, and segmentation of Couinaud anatomical segments.
Limitations
The method may struggle with nodules or veins close to the liver's edge and when nodules are similar in size to liver tissue.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website